![1
Frédéric Devaux (ed.), Yeast Functional Genomics: Methods and Protocols, Methods in Molecular Biology, vol. 1361,
DOI 10.1007/978-1-4939-3079-1_1, © Springer Science+Business Media New York 2016
Chapter 1
Using RNA-seq for Analysis of Differential Gene
Expression in Fungal Species
Can Wang, Markus S. Schröder, Stephen Hammel, and Geraldine Butler
Abstract
The ability to extract, identify and annotate large amounts of biological data is a key feature of the “omics”
era, and has led to an explosion in the amount of data available. One pivotal advance is the use of Next-
Generation Sequencing (NGS) techniques such as RNA-sequencing (RNA-seq). RNA-seq uses data from
millions of small mRNA transcripts or “reads” which are aligned to a reference genome. Comparative
transcriptomics analyses using RNA-seq can provide the researcher with a comprehensive view of the cells’
response to a given environment or stimulus.
Here, we describe the NGS techniques (based on Illumina technology) that are routinely used for
comparative transcriptome analysis of fungal species. We describe the entire process from isolation of RNA
to computational identification of differentially expressed genes. We provide instructions to allow the
beginner to implement packages in R such as Bioconductor. The methods described are not limited to
yeast, and can also be applied to other eukaryotic organisms.
Key words Candida, Next-generation sequencing, Illumina, Bioconductor
1 Introduction
Transcriptome analysis using RNA-seq quantifies transcription
levels across the entire genome [1]. Transcript numbers are mea-
sured by sequencing; the number of reads obtained from a specific
gene in a test condition compared to a control is used to measure
changes in gene expression [2, 3]. Studying the transcriptome in
different yeast species has helped elucidate the mechanisms behind
key cellular processes and pathways [4–7]. As Next-Generation
Sequencing (NGS) technologies drop in price, RNA-seq has
become widely used as a method for analyzing gene expression
under an array of conditions, and holds many advantages over
other similar analytical techniques such as microarrays [2, 8].
Although most of the techniques required can be carried out “in-
house,” there are many private companies that now provide NGS
services. They will sequence user-provided libraries, but will also](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-18-320.jpg)
![2
isolate poly(A) RNA from total RNA preparations and construct
libraries, at additional cost. This makes RNA-seq accessible to
almost any laboratory.
In this chapter, we describe how to carry out RNA-seq analysis
from RNA isolation to computational analysis. Labs without access
to next-generation sequencing technologies can use commercial
companies for the sequencing steps, and move straight to the com-
putational analysis section, where the interpretation of results is
discussed. We describe a series of tools implemented in the R sta-
tistical language [9]. A wide variety of bioinformatics tasks and
collections of R packages, such as Bioconductor [10] or CRAN
[11] make it possible to utilize R for almost any task associated
with analyzing and visualizing sequencing data. We have provided
a set of instructions that make it possible for even the beginner to
implement tools such as DeSeq2 [12] on a laptop or personal com-
puter, to analyze changes in gene expression.
2 Materials
1. NanoDrop spectrophotometer.
2. Agilent 2100 Bioanalyzer.
3. Qubit Fluorometer.
4. Dark Reader-Blue Light Transilluminator.
5. Bead beater.
6. Next-Generation DNA Sequencer (e.g., Illumina platforms
Genome Analyzer IIx, HiSeq 2500, or MiSeq), or commercial
sequencing services.
7. PC or laptop with Linux or Mac OS X as operating system and
Internet access.
1. Yeast RNA Extraction Kit (e.g., Ribopure, Ambion).
2. RNA 6000 Nano Kit (Agilent).
3. High-sensitivity DNA Kit (Agilent).
4. Zinc RNA Fragmentation Kit.
5. Gel Excision Tips (e.g., GeneCatcher).
6. PCR Purification Kits.
7. Qubit dsDNA High Sensitivity Assay Kit (or equivalent).
8. Quick Ligation Kit.
1. 1× Binding Buffer: 20 mM Tris–HCl pH 7.5, 1.0 M LiCl,
2 mM EDTA.
2. 1× Washing Buffer: 10 mM Tris–HCl pH 7.5: 0.15 M LiCl:
1 mM EDTA.
2.1 Specialized
Equipment
2.2 Kits
2.3 Buffers
and Reagents
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-19-320.jpg)
![3
3. Tris–NaCl Buffer (50 μl 1 M Tris–HCl pH 7.5, 10 μl 5 M
NaCl, and 940 μl Nuclease-free water).
4. 10 mM Tris–HCl, pH 8.5.
5. RNAlater.
6. Dynabeads Oligo (dT)25 (e.g., Dynal from Ambion).
7. dNTP mix (10 mM dATP, dTTP, dCTP, and dGTP).
8. UTP mix (10 mM dATP, dCTP, dGTP, 20 mM dUTP).
9. Reverse transcriptase with buffers (e.g., Superscript III).
10. RNaseOUT.
11. DNA Polymerase I.
12. Klenow DNA Polymerase I.
13. Klenow Fragment (3′→5′ exo-).
14. T4 DNA polymerase.
15. T4 DNA Ligase.
16. T4 polynucleotide kinase.
17. High Fidelity PCR polymerase.
18. Ribonuclease H.
19. Uracil DNA glycosylase.
20. G-50 column (e.g., Illustra Microspin).
3 Methods
1. Inoculate overnight cultures in 5 ml growth medium incubated
at a relevant temperature (often 30 °C, shaking at 200 rpm).
2. Sub-culture to an A600nm of 0.2 in 50 ml growth medium and
grow to mid-log phase (incubation time depends on growth
rates of different species being studied). At this point, an addi-
tional treatment can be used, for example treatment with a
drug, or a change in temperature or oxygen concentration.
3. Following treatment, retrieve cells from 25 ml culture either
by centrifugation for 5 min at 4 °C at 3160×g, or to avoid
stress [13] by collection on a filter (0.45 μm nitrocellulose
membrane filter) using a vacuum source.
4. Resuspend the cell pellet in 100 μl RNAlater stabilization solu-
tion and store at −80 °C until required. The RNAlater solution
inactivates any RNases and prevents any changes in expression
of the RNA.
1. Treat all lab surfaces and pipettes with 70 % ethanol and an
RNase decontamination solution (e.g., RNaseZap) to remove
any unwanted RNases.
3.1 Preparation
of RNA
3.1.1 Cell Growth
3.1.2 RNA Isolation,
Yield, and Quality
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-20-320.jpg)
![4
2. Thaw cells on ice.
3. Extract RNA using a commercial kit, following the manufac-
turer’s instructions. We use Ribopure Yeast RNA Extraction
Kit from Ambion (see Note 1).
4. Determine RNA concentrations below 50 ng/μl by measuring
absorbance with a Qubit fluorometer. Use a NanoDrop to
identify contaminants [14]. A reading at 260 nm is used to
determine concentration (A260 of 1=40 μg/ml). Proteins
absorb at 280 nm. The A260/280 ratio therefore provides a mea-
surement of the purity of the RNA; the ratio should lie between
1.8 and 2.2. Ethylenediaminetetraacetic acid (EDTA), carbo-
hydrates, and phenol all have absorbance near 230 nm. The
A260/230 ratio is therefore used as a secondary measure of nucleic
acid purity. Expected A260/230 values lie in the range of 2.0–2.2.
If the ratio is appreciably lower than this, contaminants are
probably present and the sample should not be used.
5. Measure RNA quality with a fluorometric based analytical sys-
tems, e.g., Bioanalyzer from Agilent Technologies, following
the manufacturer’s instructions. Analysis on a Bioanalyzer gen-
erates a graphical visualization in the form of an electrophero-
gram of ribosomal peaks (28S and 18S), peak ratio, RNA
concentration, a calculated RIN (RNA Integrity Number)
value, and a gel-like image of the RNA sample. The RIN value
is a measurement of the overall integrity of a given RNA sam-
ple that is not affected by sample concentration but by the
overall RNA content and background degradation. RIN values
>6 are considered to be of acceptable quality. The quantitative
range for the RNA 6000 Nano Kit is 5–500 ng/μl.
The library protocol described here was developed for sequencing
on an Illumina Genome Analyzer IIx. Most analysis is now carried
out with the more recent HiSeq and MiSeq systems from the same
company. The protocol described here may be adapted for use with
the newer platforms (HiSeq/MiSeq/NextSeq) with some minor
updates (see Subheading Adapter Synthesis). The steps required for
library generation are shown in Fig. 1. This protocol generates
strand-specific information by incorporating dUTP during the syn-
thesis of the second strand cDNA synthesis [15–17]. This is subse-
quently removed by digestion with uracil DNA glycosylase (UDG).
There are several variations of the dUTP method, including com-
bining with Illumina TruSeq kits [15, 18]. Other methods for gen-
erating strand-specific data are described by Levin et al. [16].
1. Dilute 10 μg total RNA in 50 μl using nuclease-free water.
2. Incubate at 65 °C for 5 min to disrupt RNA secondary struc-
tures, and then place on ice.
3.2 Library
Generation
3.2.1 Purification
of Poly(A) RNA (Fig. 1a)
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-21-320.jpg)
![8
The barcodes are used to separate library specific data after
sequencing [19]. We originally used home-made single read (SR)
Y-shaped adapters with short six nucleotide barcodes, designed by
Dr. Amanada Lohan UCD, and based on the 2008 Illumina cus-
tomer letter and Craig et al. [19, 20] (Fig. 2a).
These adapters are made from two single stranded oligonucle-
otides (Oligo-1 and Oligo-2) with both complementary regions
and noncomplementary regions that when annealed create a
Y-shaped adapter bound together at the hinge (complementary)
region (Fig. 2a). The top oligonucleotide (Oligo-1) contains a T
overhang with a phosphorothioate linkage required for stabiliza-
tion and resistance to nuclease digestion, that is designed to ligate
to the A overhang added to the insert DNA during library genera-
tion. Oligo-2 is phosphorylated at the 5′ end during synthesis,
and is complementary to the P7 region, which anneals to sequences
on the flowcell. An equivalent P5 region is added at the opposite
end during library amplification (step 10, Fig 1j). The 6 nucleo-
tides barcode (index) is added to both Oligo-1 and Oligo-2.
Table 1 shows six barcode indexes, allowing six libraries to be
combined in a single lane (multiplexing). The choice of barcode
depends on the number of libraries combined (Table 2). It is now
possible to design and synthesize long adapter sequences, using
updated recommendations from Illumina [21] that remove the
necessity to add the P5 region during library amplification
(Fig. 2b, [21]) (see Note 2).
1. Synthesize the oligonucleotides commercially, and using
HPLC purification. To construct the Y-shaped adapters, resus-
pend lyophilized oligonucleotides in 10 mM Tris–HCl at
100 pmol/μl and anneal them together to form a forked
adapter (iAdapter), making sure that each oligo contains the
same barcode/index sequence.
2. Add 20 μl of the relevant indexed Oligo-1, 20 μl indexed
Oligo-2 and 10 μl Tris–NaCl Buffer in 0.2 ml PCR tube.
Fig. 2 (continued) of the barcode index (using index sequence of iSR-6 (Table 1) as an example). The inserted
cDNA is shown in italics.The P7 sequence is highlighted in dark grey. Before library amplification the first (top)
strand (contain U residues) is degraded, and the bottom strand is copied using SR1.2 as a primer. Subsequent
amplification with primers SR1.1 and SR1.2 adds the P5 sequence (light grey). (b) Library generation using
updated Illumina recommendations [21].Two oligonucleotides (the universal adapter and indexed adapter) are
synthesized for each library, and a Y-shaped adapter is generated as in 2A. The cDNA is ligated at the arrow.
The universal adapter sequence contains the P5 sequence and the indexed adapter contains the P7 sequence.
The P7 sequence also contains an index/barcode (In). The libraries are amplified with primers 1 and 2. The
number of multiplexed samples can be increased by also including indexes in the universal primer (dual indexing,
not shown). For single reads a sequencing primer for the P7 end is used, for paired-end reads, primers from
both P7 and P5 ends are used. Advice on designing and synthesizing longer adapters is available from refs.
[15, 45]. The asterisk indicates a phosphorothioate linkage, and the P indicates a phosphorylated nucleotide
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-25-320.jpg)
![10
3. Incubate at 97 °C for 2 min, followed by a stepdown of
−1 °C/min for 72 cycles, and finally at 25 °C for 5 min.
4. Store the 40 μM iAdapter master stock at −20 °C. For most
RNA-seq libraries a 15 μM working stock is used. However
when <1 μg starting RNA is available further dilution may be
required.
1. Add 25 μl 2× Quick DNA ligase buffer, 1 μl iAdapter mix
(15 μM), and 2 μl Quick T4 DNA ligase (NEB) to 22 μl end-
repaired cDNA with an A overhang (from the end repair step).
2. Incubate at 20 °C for 15 min and purify by elution in 10 μl
10 mM Tris–HCl, pH 8.5 using a PCR purification kit.
3. Store at −80 °C.
1. Separate samples on a 2.5 % agarose gel by prepared using
ultrapure TAE buffer and high-resolution agarose contain-
ing Ethidium Bromide at a final concentration of 1 μg/μl
(see Note 3). Use a suitable DNA ladder (see Note 4).
3.2.7 Adapter Ligation
(Fig. 1g)
3.2.8 Gel Purification
(Fig. 1h)
Table 2
Pooling strategies
Number of libraries Best combinations
2 in the lane (iSR 6, iSR 20) or (iSR 10, iSR13)
3 in the lane (iSR 10, iSR 11, iSR13) or (iSR 6, iSR16, iSR 20)
5 in the lane (iSR 6, iSR 10, iSR 13, iSR 16, iSR 20)
6 in the lane (iSR 6, iSR 10, iSR 11, iSR 13, iSR 16, iSR 20)
Table 1
Barcode sequences
Adapter IDa
Index/barcodeb
iSR-6 AGCTAT
iSR-10 CGATCT
iSR-20 GATCGT
iSR-11 GCTAGT
iSR-13 TAGCTT
iSR-16 TCGATT
a
SR indexed adapters are designed by adding a barcode of six nucleotides, which have
to maintain color balance for each base. A/C bases are identified by the red laser and
G/T bases by the green laser on a Genome Analyzer IIx. We show six adapters designed
according to Illumina indexing guidelines [21]. Many more are possible (for example,
see ref. [45])
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-27-320.jpg)
![13
In the following sections we describe how to download, process,
and analyze RNA-seq data using Mac OS X or a Linux distribution
(such as Ubuntu) as the operating system. A server or computer
cluster (e.g., Amazon EC2) can also be used.
To illustrate the use of the software we use a subset of recently
published data from an experiment investigating the differences
between the transcriptome of Candida parapsilosis grown as bio-
films and under planktonic growth conditions (Table 3) [22].
This is strand-specific transcriptional profiling data obtained from
a commercial company (BGI, Hong Kong) using an Illumina
HiSeq 2000 with paired end reads of 90 bases. We describe in
some detail how to visualize the results using a combination of R
[9] and Bioconductor [10].
1. Download the dataset from the Gene Expression Omnibus
(GEO [23]) under GEO accession number GSE57451. It
includes wild type C. parapsilosis cultures grown under plank-
tonic and biofilm conditions. The individual reads are stored in
the Sequence Read Archive (SRA [24]) under accession num-
ber SRP041812. Download the data using your favorite tool,
or use the Unix command “wget” to import the files to your
server or hard drive. The total size is 10 GB.
2. Alternatively, all required data and output files generated by
working through the exercises in the computational analysis
section can be downloaded from http:/
/www.cgob.ie/supp_
data. The directory structure follows the Unix setup described
under “Common Unix Setup”, which makes it easy to verify
that your locally generated results are correct.
1. Use a structured directory system for all sequencing related
software and data to help to keep track of files and installed
software. The setup described here is applicable for both Mac
OS X and Linux operating systems. In Mac OS, use the
3.3.1 Download Data Set
3.3.2 Common
Unix Setup
Table 3
Sequencing Read Archive accession numbers for samples used
as an example
SRA accession number Description
SRR1278968 C. parapsilosis planktonic replicate 1
SRR1278969 C. parapsilosis planktonic replicate 2
SRR1278970 C. parapsilosis planktonic replicate 3
SRR1278971 C. parapsilosis biofilm replicate 1
SRR1278972 C. parapsilosis biofilm replicate 2
SRR1278973 C. parapsilosis biofilm replicate 3
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-30-320.jpg)
![Table 4
Overview of file formats, software, and resources used for processing and analyzing RNA-seq data
Format Description
SRA Used by the Sequence Read Archive to store and provide sequencing data
FASTQ For storing sequencing data and corresponding quality scores
GTF/GFF General Feature Format/Gene Transfer Format. Standardized formats for storing
gene information
SAM Sequence Alignment Map. Tab-delimited file format for storing alignment
information from sequencing reads
BAM Binary version of SAM format with a significantly smaller file size
BED Format to store specific meta-data for regions of the genome. Used by the UCSC
Genome Browser
BEDGRAPH Based on the BED format. Stores scored data for specified genomic regions
BIGWIG Very large collections of the BEDGRAPH format can be transformed into the
binary BIGWIG format
Software/resources Reference/website Description
SRA/SRA
Toolkit
[24] ncbi.nlm.nih.gov/Traces/sra Storage for raw sequencing reads
Collection of scripts for handling SRA files
SAMtools [37] www.htslib.org/ Collection of scripts to handle SAM files
Skewer [30] sourceforge.net/projects/skewer Tool for quality trimming and filtering of
sequencing reads
FastQC [29] bioinformatics.babraham.ac.uk/
projects/fastqc
Generates quality reports for sequencing
files
Bowtie2 [36] bowtie-bio.sourceforge.net/
bowtie2
Tool for aligning sequencing reads to a
reference genome
TopHat [31] ccb.jhu.edu/software/tophat Splice-aware aligner for sequencing reads
to a reference genome
HTSeq [38] www-huber.embl.de/users/
anders/HTSeq
Python based tools to analyze sequencing
data. The script htseq-count calculates
read counts per gene
R [9] www.r-project.org Statistical language used for a variety of
computational biology tasks
Bioconductor [10] www.bioconductor.org Large collection of R packages for
biological data
CRAN [11] cran.r-project.org Large collection of R packages
CGD [40] www.candidagenome.org Extensive resource for Candida species
IGV [25] www.broadinstitute.org/igv Integrative Genomics Viewer for displaying
sequencing data
Cytoscape [64] www.cytoscape.org Open source platform for visualizing and
analyzing network data
Python [65] www.python.org Programming language commonly used
for Bioinformatics tasks](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-33-320.jpg)
![18
8. To install HTSeq download and unpack the HTSeq archive
and install it using Python with the following commands (for
version 0.6.1):
tar xvfz HTSeq-0.6.1.tar.gz
cd HTSeq-0.6.1
python setup.py install –user
ln –s $PWD/build/scripts-2.6/htseq-count ~/local/bin
9. For Mac OS X, an installation package for R is provided.
Download the newest version, double-click the downloaded
file and follow the instructions of the Mac OS X installer. On
Linux systems execute the command:
sudo apt-get install r-base r-base-dev
If no “sudo” rights are available, download the source package
of R, unpack the archive and build R with the following
commands:
tar xvfz r-base_3.1.1.orig.tar.gz
cd R-3.1.1
./configure –-prefix=$HOME/local/bin
make
ln –s $PWD/bin/R ~/local/bin
10. The Integrative Genomics Viewer [25] can be downloaded
and used locally or launched directly from a web browser with
varying amounts of allocated memory.
11. The Bioconductor package DESeq2 is required for the differ-
ential expression analysis described under “Generating HTML
reports”. The Bioconductor package ReportingTools is used
to create HTML reports from DESeq2 results. Start R and
execute the following commands:
source("http://bioconductor.org/biocLite.R")
biocLite("DESeq2")
biocLite("ReportingTools")
Several different file formats are required. The user should become
familiar with the various types listed in Table 4 and described below.
SRA: File format used by the Sequence Read Archive [24] to store
and provide sequencing data. SRA format files can be converted
into several commonly used formats using SRA Toolkit. SRA files
in the sequence read archive format (file ending “.sra”) can be
transformed into the FASTQ format using “fastq-dump” from
SRA-tools (see Note 8).
FASTQ: A text based format for storing sequencing data and qual-
ity scores. Each entry (read) in the FASTQ format consists of
four lines. These represent (1) sequence identifier and description,
3.3.4 File Formats
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-35-320.jpg)
![19
(2) the sequence, (3) an optional line that starts with a “+” and
most commonly includes the sequence identifier and description
again and (4) the Phred scale that is used to measure the base qual-
ity. Phred scores indicate the probability of incorrect base calls and
the Phred scale is based on ASCII characters. For current Illumina
sequencing data the ASCII encoded scores have an offset of 64 and
raw base qualities normally range from character @ (quality 0) to i
(quality >40). A shortened example of one FASTQ file entry is
shown below:
1. @SRR1278968.1.1 FCC1WYWACXX:1:1101:1238:2126
length=90
2. TGGGNCTGTACGTGGTTCTTCAATTGCTTGTTTGTT
CAATGGTAAATTCG[…]
3. +SRR1278968.1.1 FCC1WYWACXX:1:1101:1238:2126
length=90
4. ___cBQaccgg^ee[ddeegghfff`gghbe_cegffaa_c^_aeedc`[…]
GTF/GFF: The General Feature Format (GFF) and Gene
Transfer Format (GTF) are two very similar formats used to
store feature (gene) information. These include the genomic
locations of exons, Coding Sequences (CDS), transcripts, 3′
and 5′ UnTranslated Regions (UTRs), tRNA, etc. The GFF
file for an organism is used to assign features to sequencing
reads that are mapped to the genome. An example of a line
from a C. parapsilosis GFF file is shown below:
Cp_c1 . exon 94585 95295 . - . gene_id "CPAR2_100565_
exon"; transcript_id "CPAR2_100565_mRNA"
It stores the chromosome name (Cp_c1), feature type (exon),
start and stop positions, strand (-), as well as additional attri-
butes that are used for feature annotation, for example the
transcript name (CPAR2_100565_mRNA). Single dots “.” in
this example indicate missing/empty information (see Note 9).
SAM: The Sequence Alignment/Map (SAM) format is a tab-
delimited file to store alignment information for sequencing
reads. There are eleven mandatory columns for each entry,
which include information such as the sequence identifier, a
bitwise FLAG that provides a summary of the read alignment,
mapping position and quality score of the mapping. Additional
columns can contain more specific information, such as the
number of times the sequence mapped to the genome (NH),
comments (CO), or mate pair information if the sequence data
is generated from paired-end reads (MC, MQ).
An example of a line from a SAM file is shown below:
HWI-D00382:125:C48G6ACXX:8:1101:1134:59125 137
Cp_c8 2005578 50 101M * 0 0 AGCTGGTATCTTGTTG
ACCCCAACTTTTGTCAAGTTGATTGCTTGGTACGATA
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-36-320.jpg)
![20
ACGAATACGGTTACTCCACCAGAGTTGTTGATTT
GTTGGAAAAATTTG CCCFFFDEHHHGHIGHHGGIID:
CGHEHHGHFGEH>HEHIGIIIHEGHIG=FHGIIG=
ACGHEHAAH;C==BBDE(.(6>A?B@;A@CACAA3(:<?B@C4
AS:i:0 XN:i:0 XM:i:0 XO:i:0 XG:i:0 NM:i:0 MD:Z:101 YT:
Z:UU XS:A:+ NH:i:1
From the beginning it lists the sequence ID, FLAG, chromo-
some name, leftmost mapping position, mapping quality,
CIGAR string for the alignment (101 matching bases),
sequence ID of mate or read pair (“*” means information
unavailable), position of mate, observed template length, raw
sequence, and Phred-scaled base quality. Information about
the optional fields, such as AS or XN, are available from
SAMtools’ GitHub repository [26].
BAM: Smaller binary version of SAM format, can be viewed
using the “samtools view” command. BAM files are commonly
used to display alignment data in genome browsers because of
their smaller size compared to the non-binary SAM format.
BED: Mostly used for displaying genomic data in a genome
browser. Three fields specify the chromosome, start and end
position. Nine additional fields can be used to provide more
specific values for the genomic location, i.e., name, score,
strand, thickStart, thickEnd, itemRgb, blockCount, block-
Sizes, and blockStarts. An example of a bed file generate from
TopHat showing deletions found in RNA-seq data compared
to the reference genome is below:
track name=deletions description="TopHat deletions"
Cp_c1 1474 1475 - 1
Cp_c1 1509 1511 - 2
Cp_c1 1771 1772 - 1
BEDGRAPH: More specific format for displaying continuous-
valued data in a genome browser. Based on the BED and WIG
formats, the BEDGRAPH format can be used to display con-
tinuous-numeric values for genomic regions, for example tran-
scriptome data. An example of a BEDGRAPH file is below.
The columns represent chromosome name, start and stop
position, and the numeric value, e.g., a user-defined score or
coverage information.
Cp_c1 665 756 -2
Cp_c1 1039 1042 -1
Cp_c1 1042 1067 -2
5. BIGWIG: For very large collections of data the BEDGRAPH
files can be converted into the binary BIGWIG format to save
disk space.
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-37-320.jpg)
![21
After successfully installing the required software listed in Table 3
on Linux or Mac OS X, all further commands can be executed in
the Terminal. Throughout the workflow below, we will provide
example commands for sample SRR1278968 (Table 3).
1. To successfully execute downstream analyses these commands
need to be executed separately for all six samples downloaded
from SRA (Table 3). This data is already stored in separate files
for each sample. If multiple samples are sequenced in the same
lane on a sequencing machine, e.g., an Illumina HiSeq 2500,
the raw sequencing reads must be separated using the bar-
code/indexing information, which is usually the first six bases
of each read. The fastx_barcode_splitter tool from the FastX-
Toolkit can be used to achieve this [27, 28].
The data in Table 3 was obtained from strand-specific 90
base paired-end sequencing. For each sample there are two
different files, one containing the first read of the pair and
one containing the second read. The standard file naming
convention for paired-end reads ends is “_1” and “_2.
Generate the files by providing fastq-dump with the option
“--split-3” as shown here:
fastq-dump --split-3 SRR1278968.sra
2. To confirm that the conversion from SRA to FASTQ was suc-
cessful, use the Unix commands “head” or “less” to briefly
examine the generated files. Each sample should have two
additional files with the endings “_1.fastq” and “_2.fastq”.
The first read in the FASTQ file looks like this:
@SRR1278968.1 FCC1WYWACXX:1:1101:1238:2126
length=90
TGGGNCTGTACGTGGTTCTTCAATTGCTTGTTTGT
TCAATGGTAAATTCGAGTCATCATGATGTGTTGGAGT
TTGATTGGTGATTGTTTG
+SRR1278968.1 FCC1WYWACXX:1:1101:1238:2126
length=90
___cBQaccgg^ee[ddeegghf f f`gghbe_cegf faa_c^_
aeedc`Xe^aeebfa]begbebbZc_bcgR`^`V^R^__]]bB
3. Once all the files are generated, check the overall quality of the
data using FastQC, a java based tool that creates extensive
summary reports. Use the following commands:
mkdir qc
fastqc SRR1278968_1.fastq -o qc 1>qc/SRR1278968_1.log
2>qc/SRR1278968_1.err
fastqc SRR1278968_2.fastq -o qc 1>qc/SRR1278968_2.log
2>qc/SRR1278968_2.err
3.4 Data Processing
3.4.1 Quality Control
and Trimming of Raw Data
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-38-320.jpg)
![22
The “mkdir” command creates the “qc” folder where the
results will be stored. For each file, FastQC generates a fastqc_
report.html file, which can be displayed in any available
browser. The quality report provides a basic summary of the
sequencing reads in each file, overall per base and per sequence
quality scores of a random subsample of all reads and several
statistics to assess the quality of the sequencing run and the
sequenced material. FastQC also reports basic sequence analy-
sis results, such as over-represented sequences and relative
kmer enrichments.
To assess of the quality of the sequencing data, it is helpful
to look at the per base sequence quality boxplot. Figure 3
shows plots from two different fastq files, one from the sample
data. The data in Fig. 3a is of a high quality. The boxplot shows
that the quality scores from the base calling rarely fall below 30
for all 90 bases in the reads. Applying quality filtering on this
sample will result in discarding a very small number of reads.
Figure 3b shows an example of a sample with reads of
mixed quality. The black boxes for the first 80 bases are close to
or above a base quality of 30, which indicates that the majority
of reads has a high quality. However, a large number of reads in
the lower 25th percentile fall below an acceptable base quality
threshold (e.g., [15]) and have to be trimmed or removed.
The quality of reads from a sequencing experiment can
vary significantly, the reasons vary from poor quality (degraded)
or contaminated starting material to mistakes during the
sequencing run itself (e.g., temporary shortage of solutions in
the sequencing machine, or bubbles in the flowcell) [29].
4. After inspecting the FastQC report, trim the raw reads using
Skewer [30] (see Note 10). Trimming sequencing data is an
important step to ensure that only high quality data is analyzed
and the results are not influence by poor quality reads. Skewer
was developed primarily to improve adapter trimming of next-
generation sequencing data, but it is also one of the fastest
tools to remove poor quality bases from paired-end RNA-seq
reads. It can utilize multiple processors to further speed up the
quality trimming [30].
To run Skewer a few options must be specified. These
include “-m pe” for paired-end trimming. Additionally recom-
mended thresholds for trimming are a minimum read length of
Fig. 3 (continued) and below each represent 25 %. Light, medium, and dark grey background colors indicate
poor, medium, and good per base quality, respectively. Quality scores are encoded in Illumina 1.5 format
(a) and >1.3 format (b). Expected quality for raw sequencing data in both formats ranges from 0 to 40, with
the exception that in Illumina format version 1.5 and above the quality score 2 represents the Read Segment
Quality Control Indicator and 0 and 1 are unused
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-39-320.jpg)
![23
40
Quality socres (Illumina 1.5)
a
b Quality socres (Illumina >1.3)
Position in read (bp)
Position in read (bp)
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
38
36
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
1 2 3 4 5 6 7 8 9 10-14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
20-24 30-34 40-44 50-54 60-64 70-74 80-84 90
Fig. 3 FastQC per base sequence quality boxplot example for high quality (a, sample SRR1278968) and low
quality data (b, plot adapted from [29]). Each black box represents 50 % of the reads and the black lines above
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-40-320.jpg)
![24
36 bases after trimming “-l 36”; a quality threshold for removing
bases of the 3′ end with scores lower than 15 “-q 15”; a thresh-
old of 15 for the mean quality of a read “-Q 15”. In the example
below the output directory for the trimmed data is “trim” and
“-t 4” specifies that four cores should be used when executing
Skewer. The two fastq files for our sample are listed at the end
of the command line, with “_1.fastq” before “_2.fastq”.
mkdir trim
skewer -m pe -l 36 -q 15 -Q 15 -o trim/SRR1278968 -t 4
SRR1278968_1.fastq SRR1278968_2.fastq
Skewer will provide a summary for each executed command,
for example for sample SRR1278968:
13722223 read pairs processed; of these:
26390 (0.19 %) short read pairs filtered out after trimming by
size control
0 (0.00 %) empty read pairs filtered out after trimming by size
control
13695833 (99.81 %) read pairs available; of these:
2902491 (21.19 %) trimmed read pairs available after
processing
10793342 (78.81 %) untrimmed read pairs available after
processing
Only 0.19 % of the reads were discarded after trimming, since
their length was shorter than 36 bases. From the other reads,
21.19 % were trimmed by a varying number of bases from the
3′ end because the base quality fell below the specified thresh-
old of 15 (“-q 15”).
5. After trimming the data, run FastQC again, this time using the
output FASTQ files from the trim folder. This ensues all the
data is of high-quality (not shown).
All RNA-seq reads must be mapped to a reference genome, using
an aligner such as TopHat [31].
1. Download the C. parapsilosis reference genome and gene
annotation [22, 32–35] from http:/
/www.cgob.ie/supp_data.
The files are called “cpar.fa” and “cpar.gff”, respectively.
2. Rename the files generated by Skewer and create an index for the
reference genome. For paired-end reads, TopHat requires that
the FASTQ files end in “_1.fastq” and “_2.fastq”, for the first
and second mate respectively. For single-end reads no naming
convention or order of samples exists. Renaming the trimmed
FASTQ files that currently end with “pair1.fastq” and “pair2.
fastq” is easily achieved using the Unix rename command:
rename 's/pair/pair_/' *pair*
3.4.2 Mapping Reads
to the Genome
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-41-320.jpg)
![25
3. To create the reference genome index use bowtie2-build [36]
with the FASTA genome file and output folder “cpar-index”:
mkdir cpar-index
bowtie2-build cpar.fsa cpar-index/cpar
4. Execute TopHat with the following command:
tophat -p 12 -o SRR1278968 -G cpar.gff -g 1 --b2-very-
sensitive
--library-type fr-firststrand cpar-index/cpar
trim/SRR1278968-pair_1.fastq trim/SRR1278968-pair_2.
fastq
The option “-p” sets the number of processing cores TopHat
utilizes, “-o” sets the output folder, “-G” is optional and pro-
vides genome annotation and “-g 1” sets the maximum amount
of times a read can map to the genome before it is reported as
ambiguously mapped.
The preset “--b2-very-sensitive” is specified, which
includes a number of settings (-D 20 -R 3 -N 0 -L 20 -i
S,1,0.50). The D and R options specify “effort” options of
TopHat. The higher these numbers are, the higher the amount
of attempts TopHat will execute to realign reads or extend
existing alignments. The N, L and i options fine-tune how
TopHat tries to align the reads. The number of mismatches
that are allowed during seed alignment (N), the length of the
seed substring (L) and the function for the interval between
substrings (i). Further information on TopHat can be found in
the online manual (follow link in Table 4).
The very-sensitive option is used to increase the probabil-
ity of mapping reads, as well as the length of the alignment. If
the full read does not map to the reference genome, it is cut
into smaller pieces (seeds) that TopHat tries to realign.
To specify that the RNA-seq data is strand-specific, set the
library-type option for TopHat (“--library-type fr-firststrand”)
(see Note 11).
The C. parapsilosis genome, like many other eukaryotes,
includes several introns [34, 35]. For this reason, a splice-
aware aligner is used to map reads to the reference genome.
TopHat has this ability, as do other tools (see Note 12).
Aligning the reads generated several files. The most impor-
tant one is “accepted_hits.bam”, which includes all reads that
were successfully mapped to the genome, as well as the map-
ping quality and the mapping position. The “unmapped.bam”
file lists all reads that were not successfully mapped to the
genome. The files “deletions.bed” and “insertions.bed” show
positions where reads were successfully mapped to the genome,
but compared to the reference genome bases were either miss-
ing in the read (included in “deletions.bed”) or additional
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-42-320.jpg)
![26
bases were present in the read (“insertions.bed”). The file
“junctions.bed” lists all positions in the genome where reads
would span a region. This includes regions where reads span an
intron. Visualizing the junctions in a genome browser can help
identify different isoforms of a gene.
5. To check the data for properly aligned mate pairs, generate a
summary of the alignment file from TopHat using the flagstat
script from SAMtools [37] with the following command:
samtools flagstat accepted_hits.bam
The summary lists the total number of reads with a detailed
breakdown of the paired reads. An example output is shown
below. Here, 97.25 % of aligned reads are properly paired and
only a very minor subset of reads without a mate or with a
mate mapping to a different chromosome are present.
25680571 + 0 in total (QC-passed reads + QC-failed reads)
0 + 0 duplicates
25680571 + 0 mapped (100.00%:-nan%)
25680571 + 0 paired in sequencing
12793758 + 0 read1
12886813 + 0 read2
24974160 + 0 properly paired (97.25%:-nan%)
25135208 + 0 with itself and mate mapped
545363 + 0 singletons (2.12%:-nan%)
41182 + 0 with mate mapped to a different chr
41182 + 0 with mate mapped to a different chr (mapQ>=5)
6. When the reads are mapped to the genome, prepare the aligned
“.bam” files for further analysis and viewing in a genome
browser. Visualizing the reads at this point in the analysis is a
good way to identify any potential problems, such as incorrect
mapping of paired mates, or incorrect orientation of strand-
specific data.
First index the “accepted_hits.bam” file. This is essential for
genome browsers to display reads efficiently. The following
SAMtools command generates a “.bai” file, which contains the
bam index.
samtools index accepted_hits.bam accepted_hits.bam.bai
1. Install the IGV browser as described in “Installing Software”
(see Note 13).
2. Load the reference genome sequence and the annotation. This
is achieved from a single dialog box under “Genomes -> Create
.genome File…”.
3.4.3 Visualizing
the Mapped Reads
in a Genome Browser
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-43-320.jpg)
![27
3. Enter a unique identifier for the genome, and select the fasta
file that was used earlier for building the Bowtie2 genome index
as the “FASTA file”. Alternatively, select the genome annota-
tion GFF file that was used with TopHat as the “Gene file”.
4. To load BAM files into IGV, select the reference genome from
the drop-down menu at the top left corner of IGV.
5. Select a BAM file generated by an aligner from the local file
system, a server or URL. The accompanying BAM index
file must be in the same directory as the BAM file itself.
6. Load the GFF file to display the annotation. Initially, no
sequencing reads are visible. This is because the default view in
IGV is to show the entire chromosome, and displaying all reads
mapped to the chromosome requires too much memory.
Sequencing reads will be displayed if the visible region of the
chromosome is set to below 100 kb in length. A snapshot of
IGV with a BAM file and genome annotation loaded is shown
in Fig. 4.
7. The reads can be displayed in three different ways: collapsed,
squished and expanded. This is selected by right-click on the
“accepted_hits.bam” label on the left side of the track. It also
can be helpful to visualize paired-end RNA-seq data. To dis-
play read pairs as connected reads, select “View as pairs” again
by right-click on the “accepted_hits.bam” label. By default
IGV will display read pairs in different colors since the reads
have different directions. To adjust the coloring schema to
Fig. 4 Snapshot of IGV showing strand-specific C. parapsilosis RNA-seq data [22]. Included are RNA-seq cov-
erage (top track), BAM file (middle track) and genome annotation (bottom track). Reads on the forward strand
are dark grey and light grey on the reverse strand. Arrows inside the annotation track indicate the direction
transcription.The data range of the RNA-seq coverage is indicated in square brackets [0-4135]. BAM file reads
are displayed using the “squished” visualization option and colored using the “first-of-pair strand” option
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-44-320.jpg)
![28
show transcriptional orientation, right-click into the “accepted_
hits.bam” label again and choose “Color alignments by” ->
“first-of-pair-strand”.
To measure transcripts, the number of mapped reads for each gene
must be counted. The Python script htseq-count from HTSeq
[38] does exactly this. However, in order to run htseq-count,
mapped reads must first be sorted in the bam file according to their
location on the genome, and the file converted into the non-binary
and significantly larger SAM format.
1. Sort the reads by location (option “-n”) and convert the sorted
BAM file to the SAM format using the following two SAMtool
commands. A descriptive header for the SAM file is included
with the option “-h”.
samtools sort -n accepted_hits.bam accepted_hits.sorted
samtools view -h -o accepted_hits.sorted.sam accepted_hits.
sorted.bam
2. To run htseq-count, specify the following command for each
sample:
htseq-count -m union -s reverse -t exon -i transcript_id
-o accepted_hits.sorted.sam.htseq accepted_hits.sorted.sam
../cpar.gff 1>accepted_hits.sorted.sam.htseq.count
2>accepted_hits.sorted.sam.htseq.count.log
The option “–m union” specifies how the HTSeq algorithm
assigns a read to a gene (also referred to as “feature”). The
union option is recommended in most cases [38]. The strand
direction (-s), the feature type (-t, third column in the GFF
file), and the attribute for that htseq-count should report the
read counts (-i), e.g., for each exon, or for each transcript,
should also be specified.
After successfully executing htseq-count the count data will
be stored in “accepted_hits.sorted.sam.htseq.count”. This is a
tab-delimited file with transcript names in the first column and
read counts in the second:
CPAR2_100010_mRNA 271
CPAR2_100020_mRNA 454
CPAR2_100030_mRNA 3277
In the following section we will explain how to identify differ-
entially expressed genes from read count data and subsequently
uncover the biological differences between different conditions.
Identify differentially expressed genes from read count data using the
Bioconductor package DESeq2, which assumes that the read count
data follows a negative binomial distribution [12]. (see Note 14).
3.4.4 Counting
Transcripts
3.4.5 Differential Gene
Expression Analysis
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-45-320.jpg)
![29
Run Packages in R from the R command line or from the
graphical user interface. (see Note 15).
1. Once R and DESeq2 are installed (see “Installing Software” for
instructions) open R and load DESeq2 and ReportingTools
[39] into the R environment using the following commands:
library("DESeq2")
library("ReportingTools")
2. Set the working directory to the analysis folder, for example:
setwd("~/ngs/data/")
3. The count data are stored inside each TopHat output folder as
specified for the TopHat command in “Mapping reads to the
genome”. To read these into R, use the function read.table():
samples <- c("SRR1278968","SRR1278969","SRR1278970",
"SRR1278971","SRR1278972","SRR1278973")
cDataAll <- NULL
for(i in 1:length(samples)){
file <- read.table(
sprintf("%s/accepted_hits.sorted.sam.htseq.count",
samples[i]))
cDataAll <- cbind(cDataAll, file[,2])
}
rownames(cDataAll) <- file[,1]
colnames(cDataAll) <- samples
4. The count data from all six samples is now stored in the
“cDataAll” variable. The example data (Table 3) includes
measurements from three planktonic and three biofilm sam-
ples. To specify the conditions for each sample create the vari-
able “groups” with “P” for planktonic and “B” for biofilm
(see Note 16):
groups <- factor(x=c(rep("P", 3), rep("B", 3)), levels=c("P",
"B"))
5. The htseq-count data from Subheading “Counting transcripts”
includes additional rows that indicate how many reads could
not be associated with a unique feature, or where the align-
ment quality was too low. Exclude these five rows from the
analysis using the match() function:
cData <- cDataAll[-match(x=c("__no_feature", "__ambiguous",
"__too_low_aQual", "__not_aligned", "__alignment_not_
unique"), table=rownames(cDataAll)),]
6. DESeq2 uses raw count data as input, because it normalizes
the data internally. To get a compact overview plot of all count
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-46-320.jpg)
![30
data, use Tags Per Million (TPM) normalization and the
density function:
tpm <- t(t(cData)/colSums(cData))*1e6
inlog <- log(tpm)
colLabel <- c(rep("#E41A1C", 3), rep("#377EB8", 3))
colTy <- c(rep(1:3, 3), rep(1:3, 3))
plot(density(inlog[,1]), ylim=c(0,0.4), main="Density plot of
counts per gene", lty=colTy[1], xlab="Log of TPM per gene",
ylab="Density", col=colLabel[1])
for(i in 2:ncol(tpm)){
lines(density(inlog[,i]), lty=colTy[i], col=colLabel[i])
}
legend("topright", legend=colnames(tpm), lty=colTy, col=
colLabel)
This generates a density plot that shows the distribution of the
log transformed TPM count data for each sample. It should
follow a negative binomial distribution and with maximum
log(TPM) between 4 and 5. All samples should have a very
similar distribution. If this is not the case for any one sample, it
is an early indicator that the transcriptome is very different to
other samples. This could be due to a number of reasons. If no
quality issues were detected in the raw sequencing reads and
the mapping frequency to the reference genome was above
95 %, it may indicate that there is a problem with the biological
sample.
7. The Principal Coordinate Analysis (PCoA) plot can also be
used to characterize how similar samples are. PCoA returns
coordinates that represent the dissimilarities between samples
as distances. Use the normal plot() function to create the
PCoA plot:
d <- dist(t(tpm))
fit=cmdscale(d, eig=TRUE, k=2)
x=fit$points[,1]
y=fit$points[,2]
plot(x, y, type="p", pch=20)
text(x, y, labels=row.names(t(tpm)), cex=1, adj=c(-0.25,-0.25))
Figure 5 shows the PCoA plot using data from Table 3. Control
and test samples should cluster separately and samples within
each group should cluster together. The x-axis in Fig. 5 (PCoA
dimension 1) clearly separates the control and test samples and
the y-axis (PCoA dimension 2) indicates small differences
within each group, but with a 10× smaller scale.
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-47-320.jpg)
![31
8. When you are confident that the data does not have any bias or
other confounding factors, carry out differential expression
analysis. To keep structure in the analysis directory, create the
folder DESeq2 and set it as the working directory.
if (file.exists("DESeq2")){
setwd("DESeq2")
} else {
dir.create("DESeq2")
setwd("DESeq2")
}
9. Use the following script to run DESeq2:
colData <- DataFrame(condition=groups)
dds <- DESeqDataSetFromMatrix(cData, colData, formula
(~condition))
dds <- DESeq(dds)
res <- results(dds, cooksCutoff=FALSE)
−40000 −20000 0 20000
−4000
−2000
0
2000
4000
6000
Dimension 1
Dimension
2
P3
P2
P1
B3
B2
B1
Fig. 5 PCoA plot showing transcriptional profiles of C. parapsilosis cells grown in
planktonicconditions(P1-P3,SRAIDs:SRR1287968,SRR1278969,SRR1278970)
and biofilm conditions (B1-B3, SRA IDs: SRR1278971, SRR1278972,
SRR1278973) [22].The two groups are visually separated by Dimension 1.There
is minor variation among the biological replicates, which is indicated by the ten-
fold smaller scale for Dimension 2 compared to Dimension 1
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-48-320.jpg)
![32
The Data.Frame “colData” contains the group factor that
DESeq2 uses to separate the samples. The DESeqDataSet
FromMatrix() function takes the count data, sample groups,
and the formula for comparing the samples as input and creates
a DESeqDataSet object, which can be used to run DESeq2.
The function DESeq() executes DESeq and writes all results
into the DESeqDataSet object. Use the results() function to
create a Data.Frame containing the results, such as gene name,
log2 fold change, and adjusted p-value.
10. To write the results to a file that can be opened with Excel use
the write.csv() function. In addition the result Data.Frame can
be ranked by the log2 fold change to enable easier analysis of
the data.
res <- res[order(res$log2FoldChange, decreasing=TRUE),]
write.csv(as.data.frame(res), file="p-vs-b_results.csv")
Below are the first rows of the results from the CSV file.
geneID baseMean log2FoldChange lfcSE stat pvalue padj
CPAR2_203270_
mRNA
5478.47 11.33 0.42 26.44 3.91e−154 7.59e−152
CPAR2_807700_
mRNA
20856.65 9.68 0.17 54.60 0 0
11. To extract the numbers of significantly upregulated and down-
regulated genes you must specify the results thresholds. We
recommend using a log2 fold change greater than 1 for genes
with increased expression or lower than −1 for genes with
decreased expression. Set the significant adjusted p-value
(padj) to less than 0.01.
up <- rownames(res[!is.na(res$padj) & res$padj <= 0.01 &
res$log2FoldChange >= 1, ])
down <- rownames(res[!is.na(res$padj) & res$padj <= 0.01 &
res$log2FoldChange <= -1, ])
sprintf("%s genes up-regulated, %s genes down-regulated",
length(up), length(down))
12. To enable downstream analysis of the differentially expressed
genes, save the gene IDs in two files, called “p-vs-b_up-
regulated.txt” and “p-vs-b_down-regulated.txt” with the com-
mand below. In the GFF file the gene IDs contain the ending
“_mRNA”, which is removed using the R function sub() with
the pattern “_mRNA” and an empty replacement “”.
write.table(sub(pattern = "_mRNA", replacement = "", x = up),
file="p-vs-b_up-regulated.txt", col.names=FALSE,
row.names=FALSE, quote = FALSE)
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-49-320.jpg)
![33
write.table(sub(pattern = "_mRNA", replacement = "", x =
down),
file="p-vs-b_down-regulated.txt", col.names=FALSE,
row.names=FALSE, quote = FALSE)
13. The DESeq2 package contains several methods to create over-
view plots of the differentially expressed genes. Use MA plot
(plotMA(dds)) to create a plot of the log2 fold changes against
mean normalized counts for each gene. Use the plotPCA
function to create a Principal Component Analysis (PCA) plot.
The number of genes that are taken into account for the dis-
tance calculation of the PCA can be specified. With “ntop=500”
the 500 genes with the highest row variance, i.e., the highest
variance of read counts per gene across all samples, will be
used for the PCA.
plotPCA(dds, ntop=500)
14. To represent the log2 fold change distribution of all genes as a
histogram highlighting differentially expressed genes for exam-
ple in grey (as shown in Fig. 6), use the commands:
hist(res[!is.na(res$padj) & res$padj <= 0.01, ][,"log2Fold
Change"],
breaks=seq(-15,15,0.25),
col=c(rep("tomato", 56), rep("white", 8), rep("tomato", 56)),
xlab="Log2 Fold Change", main="Overall Log2 fold change,
p.adj<0.01") (see Note 17).
Log2 Fold Change
Frequency
−15 −10 −5 0 5 10 15
0
50
100
150
200
250
300
350
Fig. 6 Histogram of log2 fold changes obtained by comparing the transcriptional profiles of C. parapsilosis cells
grown in planktonic vs. biofilm conditions [22] using the Bioconductor package DESeq2. Significant log2 fold
changes are colored in dark grey (greater than 1, or less than −1), not significant log2 fold changes are in white
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-50-320.jpg)
![34
As a last step, searchable HTML reports of the differentially
expressed genes can easily be generated using the Bioconductor
package ReportingTools [39].
1. Load ReportingTools into the R environment.
library("ReportingTools")
2. Use the functions HTMLReport() and publish() to generate a
website from the results data.frame created by DESeq2.
htmlRep <- HTMLReport(shortName="P-vs-B_results",
reportDirectory = "./reports")
publish(cbind(GeneID=rownames(res),as.data.frame(res)),
htmlRep)
3. Finally, create the report.
finish(htmlRep)
1. After the list of significantly differentially expressed genes is
generated, several tools and websites can help to identify the
biological mechanisms that drive the change between the two
conditions, planktonic and biofilm transcriptomes in this
example.
Meta-information for Candida species is available from
the Candida Genome Database (CGD, [40]). Equivalent sites
for other genomes include Saccharomyces species from the
Saccharomyces Genome Database (SGD, [41]) and Aspergillus
species from the Aspergillus Genome Database (AspGD, [42]).
For less characterized species, working with homologs from a
closely related species can provide more biological insight.
Homology information for Candida and Saccharomyces species
is available from the Candida Gene Order Browser (CGOB,
[35, 43]) and the Yeast Gene Order Browser (YGOB, [44]), as
well as from AspGD.
2. Gene Ontology (GO) Analysis: Using the identified differen-
tially expressed genes from the C. parapsilosis planktonic and
biofilm samples, Gene Ontology terms that are enriched in
upregulated or downregulated genes are identified using the
GO Term Finder at CGD ([40], Table 4). In step 1 on the GO
Term Finder website, choose Candida parapsilosis as the target
species. For step 2 upload the file “p-vs-b_up-regulated.txt”
or the equivalent file containing downregulated genes, which
were saved at the end of the DESeq2 workflow. Choose one of
the three Gene Ontologies in step 3, i.e., Biological Process,
Molecular Function or Cellular Component. We want to know
if our upregulated genes have any enriched GO terms in the
Molecular Function ontology. Fig. 7 shows a screenshot of the
GO Term Finder with sample settings. To start the analysis,
click on the Search-button in the bottom left corner.
3.4.6 Generating HTML
Reports
3.4.7 Downstream
Analysis of Differentially
Expressed Genes
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-51-320.jpg)
![36
[21] that remove the necessity to add the P5 region during
library amplification (Fig 2b [21]). The P5 sequence is included
in the first oligonucleotide (the universal adapter). The index-
ing sequence is contained in the second oligonucleotide
(indexed adapter), outside the short region that anneals with
the universal adapter (Fig 2b). The sequence of twenty-seven 6
nucleotide indexes are provided by Illumina ([21], Oligo-
nucleotide sequences, 2007–2013 Illumina, Inc. All rights
reserved) and other home-made designs are described by Ford
et al. [45]. Multiplexing of samples can be increased by also
including an index sequence in the universal adapter (dual
indexing). Recent kits from Illumina use six or eight nucleo-
tide indexes, with up to eight different versions of the universal
adapter, and 48 of the indexed adapter [21]. The libraries are
amplified using regions derived from the P5 and P7 regions,
and can be sequenced from either end using two sequencing
primers. These adapters can also be combined with dUTP
methodology for strand-specific sequencing [18].
3. AMPure XP beads (Beckman Coulter) are now recommended
for clean-up procedures instead of gel purification.
4. We use 2-log DNA ladder from NEB, which allows visualiza-
tion of DNA bands for 0.1–10 kb. Make 2-log DNA ladder
mix by mixing 1 μl 2-log DNA ladder (NEB), 1 μl Blue load-
ing dye (Promega), and 4 μl distilled water.
5. Consideration should be given to the length of the reads
generated, and whether single-end or paired-end sequences
are used. The adapters described here are for single read only,
different sequences are required for paired end reads. It is pos-
sible to obtain increasingly long reads, but at a price. Long
reads (>100 bases) are not necessary for RNA-seq. Paired-end
reads can help in mapping, but are not strictly necessary.
Strand-specific information however is strongly recommended
as it enables identification of UTRs (untranslated regions) and
antisense expression.
6. For users who are not yet proficient with the Terminal com-
mand line, a detailed Unix tutorial is available [46]. Commands
and techniques described in the tutorial are applicable for both
Linux and Mac OS X users. A large community of researchers
that work with next-generation sequencing data can be reached
on the SEQanswers forum [47]. BioStar [47, 48] is a more
general and also very useful Bioinformatics forum.
7. If any error messages arise during the installation of Python, or
if there are general questions about Unix environment vari-
ables or Unix commands, Stack Overflow [49] is a useful start-
ing point to search for solutions.
8. More information on SRA formats are available in the SRA
Knowledge Base [50] and the SRA Handbook [51].
Can Wang et al.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-53-320.jpg)
![37
9. More information about the GFF format is available at a dedi-
cated website from the Sanger Institute [52].
10. There are several equivalently suitable tools available for trim-
ming sequencing reads, some with extensive options to specify
quality thresholds and filtering parameters. It is however for
the user to decide which tool fits best for a specific task. The
tool used in this RNA-seq workflow is Skewer [30]. Other
popular tools include fastx_trimmer [27, 28], cutadapt [53],
and TrimGalore! [54]. These tools are all capable of trimming
sequencing reads based on quality thresholds.
11. There are different methods for generating strand-specific
RNA-seq libraries [16]. To verify that the correct option for
the RNA-seq data was used in TopHat, compare the aligned
reads in a genome browser with the HTSeq count data for a
specific gene. For unstranded RNA-seq data specify
“fr-unstranded”.
12. Other splice-aware aligners include GSNAP [55] or STAR
[56]. STAR is a recently developed alignment tool that has
significant speed advantages over other aligners. This is help-
ful for identifying the optimal parameters for aligning sequenc-
ing reads to a reference genome, e.g., number of allowed
mismatches per read, number of times a read is allowed to
map to the genome, or the length limitations for introns. A
comparison of the different aligners was carried out by
Engstrom et al. [57].
13. Other popular genome browsers are Artemis [58], the
Integrative Genomics Browser [25], the web-browser based
genome browsers JBrowse [59] and the UCSC Genome
Browser [60].
14. Selecting a statistical package to identify differentially expressed
genes from gene count data is an important step in an RNA-
seq workflow. There is however not one method that is supe-
rior to all others. Commonly used packages include DESeq2,
edgeR, baySeq, DEGSeq, NOISeq, tweeDEseq, and many
more. The Bioconductor version 2.14 lists 138 packages used
for Differential Expression analysis. DESeq2, edgeR, baySeq,
and EBSeq assume that the count data follows a negative bino-
mial distribution, which is the most commonly assumed distri-
bution for RNA-seq data. Choosing any of the available
methods will yield results, more important than the method
itself however is that experiments are planned with enough
replicates and that options recommended by the authors of
each method are used. It is not ideal to switch between statisti-
cal methods when comparing different datasets, which could
introduce additional biases. Detailed comparisons and analyses
of several statistical methods for identifying differentially
expressed genes from count data can be found at [61, 62].
Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-54-320.jpg)
![declared that he would sooner or later have him, the Talmud, put to
death by the hangman![48]
For the benefit of the average reader as well as to illuminate the
general subject, a short description of the Talmud will be given.
Definition.—Many attempts have been made to define the Talmud,
but all definition of this monumental literary production is necessarily
inaccurate and incomplete because of the vastness and peculiarity of
the matter treated. To describe it as an encyclopedia of the life and
literature, law and religion, art and science of the Hebrew people
during a thousand years would convey only an approximately correct
idea of its true meaning, for it is even more than the foregoing
descriptive terms would indicate. Emanuel Deutsch in his brilliant
essay on the Talmud defines it as "a Corpus Juris, an encyclopedia of
law, civil and penal, ecclesiastical and international, human and
divine. It is a microcosm, embracing, even as does the Bible, heaven
and earth. It is as if all the prose and poetry, the science, the faith
and speculation of the Old World were, though only in faint
reflections, bound up in it in nuce."
Benny describes it as "the Talmud—that much maligned and even
more misunderstood compilation of the rabbins; that digest of what
Carlyle would term allerlei-wissenschaften; which is at once the
compendium of their literature, the storehouse of their tradition, the
exponent of their faith, the record of their acquirements, the
handbook of their ceremonials and the summary of their legal code,
civil and penal."
To speak of the Talmud as a book would be inaccurate. It is a small
library, or collection of books. "Modern editions of the Talmud,
including the most important commentaries, consist of about 3,000
folio sheets, or 12,000 folio pages of closely printed matter, generally
divided into twelve or twenty volumes. One page of Talmudic
Hebrew intelligibly translated into English would cover three pages;
the translation of the whole Talmud with its commentaries would](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-56-320.jpg)
![accordingly make a library of 400 volumes, each numbering 360
octavo pages."[49]
It would be well to bear in mind that the contents of the Talmud
were not proclaimed to the world by any executive, legislative, or
judicial body; that they were not the result of any resolution or
mandate of any congregation, college, or Sanhedrin; that they were
not, in any case, formal or statutory. They were simply a great mass
of traditionary matter and commentary transmitted orally through
many centuries before being finally reduced to writing. Rabbinism
claims for these traditions a remote antiquity, declaring them to be
coeval with the proclamation of the Decalogue. Many learned
doctors among the Jews ascribe this antiquity to the whole mass of
traditional laws. Others maintain that only the principles upon which
Rabbinic interpretation and discussion are based, can be traced back
so far. But it is certain that distinct traditions are to be found at a
very early period in the history of the children of Israel, and that on
their return from Babylonian captivity these traditions were delivered
to them by Ezra and his coadjutors of the Great Assembly.
This development of Hebrew jurisprudence along lines of written and
oral law, Pentateuch and Talmud, Mosaic ordinance and time-
honored tradition, seems to have followed in obedience to a general
principle of juristic growth. Lex scripta and lex non scripta are
classical Roman terms of universal application in systems of
enlightened jurisprudence. A charter, a parchment, a marble column,
a table of stone, a sacred book, containing written maxims defining
legal rights and wrongs are the beginnings of all civilized schemes of
justice. Around these written, fundamental laws grow and cluster the
race traditions of a people which attach themselves to and become
inseparable from the prime organic structure. These oral traditions
are the natural and necessary products of a nation's growth and
progress. The laws of the Medes and Persians, at once unalterable
and irrevocable, represent a strange and painful anomaly in the
jurisprudence of mankind. No written constitution, incapable of
amendment and subject to strict construction, can long survive the](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-57-320.jpg)
![growth and expansion of a great and progressive people. The ever-
changing, perpetually evolving forms of social, commercial, political,
and religious life of a restless, marching, ambitious race, necessitate
corresponding changes and evolutions in laws and constitutions.
These necessary legal supplements are as varied in origin as are the
nations that produce them. Magna Charta, wrung from John at
Runnymede, became the written basis of English law and freedom,
and around it grew up those customs and traditions that—born on
the shores of the German Ocean, transplanted to the Isles of Britain,
nurtured and developed through a thousand years of judicial
interpretation and application—became the great basic structure of
the Common Law of England.
What the Mosaic Code was to the ancient Hebrews, what Magna
Charta is to Englishmen, the Koran is to Mahometans: the written
charter of their faith and law. Surrounding the Koran are many
volumes of tradition, made up of the sayings of Mahomet, which are
regarded as equally sacred and authoritative as the Koran itself.
These volumes of Mahometan tradition are called the Sonna and
correspond to the Talmud of the Hebrews. An analysis of any great
system of jurisprudence will reveal the same natural arrangement of
written and oral law as that represented by the Pentateuch and the
Talmud of the Jews.
The word "Talmud" has various meanings, as it appears in Hebrew
traditional literature. It is an old scholastic term, and "is a noun
formed from the verb 'limmed'='to teach.' It therefore means,
primarily, 'teaching,' although it denotes also 'learning'; it is
employed in this latter sense with special reference to the Torah, the
terms 'Talmud' and 'Torah' being usually combined to indicate the
study of the Law, both in its wider and its more restricted sense."[50]
It is thus frequently used in the sense of the word "exegesis,"
meaning Biblical exposition or interpretation. But with the
etymological and restricted, we are not so much interested as with
the popular and general signification of the term "Talmud." Popularly
used, it means simply a small collection of books represented by two](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-58-320.jpg)
![distinct editions handed down to posterity by the Palestinian and
Babylonian schools during the early centuries of the Christian era.
Divisions of the Talmud.—The Talmud is divided into two component
parts: the Mishna, which may be described as the text; and the
Gemara, which may be termed the commentary.[51] The Mishna,
meaning tradition, is almost wholly law. It was, indeed, of old,
translated as the Second or Oral Law—the δευτέρωσιϛ—to
distinguish it from the Written Law delivered by God to Moses. The
relationship between the Mishna, meaning oral law, and the Gemara,
meaning commentary, may be illustrated by a bill introduced into
Congress and the debates which follow. In a general way, the bill
corresponds to the Mishna, and the debates to the Gemara. The
distinction, however, is that the law resulting from the passage of
the bill is the effect and culmination of the debate; while the Mishna
was already law when the Gemara or commentary was made.
As we have seen above, Hebrew jurisprudence in its principles and
in the manner of their interpretation was chiefly transmitted by the
living voice of tradition. These laws were easily and safely handed
down from father to son through successive generations as long as
Jewish nationality continued and the Temple at Jerusalem still stood.
But, with the destruction of the Temple and the banishment of the
Jews from Palestine (A.D. 70), the danger became imminent that in
the loss of their nationality would also be buried the remembrance of
their laws. Moved with pity and compassion for the sad condition of
his people, Judah the Holy, called Rabbi for preëminence, resolved to
collect and perpetuate for them in writing their time-honored
traditions. His work received the name Mishna, the same which we
have discussed above. But it must not be imagined that this work
was the sudden or exclusive effort of Rabbi Judah. His achievement
was merely the sum total and culmination of the labors of a long line
of celebrated Hebrew sages. "The Oral Law had been recognized by
Ezra; had become important in the days of the Maccabees; had been
supported by Pharisaism; narrowed by the school of Shammai,
codified by the school of Hillel, systematized by R. Akiba, placed on](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-59-320.jpg)
![a logical basis by R. Ishmael, exegetically amplified by R. Eliezer, and
constantly enriched by successive rabbis and their schools. Rabbi
Judah put the coping-stone to the immense structure."[52]
Emanuel Deutsch gives the following subdivisions of the Mishna:
The Mishna is divided into six sections. These are subdivided
again into 11, 12, 7, 9 (or 10), 11, and 12 chapters,
respectively, which are further broken up into 524
paragraphs. We shall briefly describe their contents:
Section I. Seeds: of Agrarian Laws, commencing with a
chapter on Prayers. In this section, the various tithes and
donations due to the Priests, the Levites, and the poor, from
the products of the lands, and further the Sabbatical year and
the prohibited mixtures in plants, animals, garments, are
treated of.
Section II. Feasts: of Sabbaths, Feast, and Fast days, the
work prohibited, the ceremonies ordained, the sacrifices to be
offered, on them. Special chapters are devoted to the Feast of
the Exodus from Egypt, to the New Year's Day, to the Day of
Atonement (one of the most impressive portions of the whole
book), to the Feast of Tabernacles and to that of Haman.
Section III. Women: of betrothal, marriage, divorce, etc., also
of vows.
Section IV. Damages: including a great part of the civil and
criminal law. It treats of the law of trover, of buying and
selling, and the ordinary monetary transactions. Further, of
the greatest crime known to the law, viz., idolatry. Next of
witnesses, of oaths, of legal punishments, and of the
Sanhedrin itself. This section concludes with the so-called
"Sentences of the Fathers," containing some of the sublimest
ethical dicta known in the history of religious philosophy.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-60-320.jpg)
![Section V. Sacred Things: of sacrifices, the first-born, etc.;
also of the measurements of the Temple (Middoth).
Section VI. Purifications: of the various levitical and other
hygienic laws, of impure things and persons, their
purification, etc.[53]
Recensions.—The Talmud exists in two recensions: the Jerusalem
and the Babylonian. These two editions represent a double Gemara;
the first (Jerusalem) being an expression of the schools in Palestine
and redacted at Tiberias about 390 A.D.; the second (Babylonian)
being an expression of the schools in Babylonia and redacted about
365-427 A.D.
The Mishna, having been formed into a code, became in its turn
what the Pentateuch had been before it, a basis of discussion and
development. The Gemara of the Jerusalem Talmud embodies the
critical discussions and disquisitions on the Mishna by hundreds of
learned doctors who lived in Palestine, chiefly in Galilee, from the
end of the second till about the middle of the fifth century of the
Christian era. The Gemara of the Babylonian Talmud embodies the
criticisms and dissertations on the same Mishna of numerous learned
doctors living in various places in Babylonia, but chiefly those of the
two great schools of Sura and Pumbaditha.[54] The Babylonian
Talmud is written in "West Aramæan," is the product of six or seven
generations of constant development, and is about four times as
large as that of the Jerusalem Talmud, which is written in "East
Aramæan."[55] It should be kept clearly before the mind that the
only difference between these two recensions is in the matter of
commentary. The two sets of doctors whose different commentaries
distinguish the two Talmuds dealt with the same Mishna as a basis of
criticism. But decided differences are noticeable in the subject
matter and style of the two Gemaras represented by the two
recensions of the Talmud. The discussions and commentaries in the
Jerusalem Talmud are simple, brief, and pointed; while those of the
Babylonian Talmud are generally subtle, abstruse, and prolix. The
dissertations in the Jerusalem Talmud are filled to overflowing with](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-61-320.jpg)
![consider that the Mishna has a separate existence from the Talmud
and a distinct recension of its own. In this state it is simply a naked
code of laws. But when the Gemara has been added to it the Talmud
is the result, which, in its turn, becomes a distinct entity and may be
referred to as such in the same sentence with the Mishna.
Relation of Talmud to Pentateuch.—As before suggested, the
Pentateuch, or Mosaic Code, was the Written Law and the very
foundation of ancient Hebrew jurisprudence. The Talmud, composed
of the Mishna, i.e., Tradition, and the Gemara, i.e., Commentary, was
the Oral Law, connected with, derived from, and built upon the
Written Law. It must be remembered that the commonwealth of the
Jews was a pure theocracy and that all law as well as all religion
emanated directly or indirectly from Jehovah. This was as true of
Talmudic tradition as of Mosaic ordinance. Hillel, who interpreted
tradition, was as much inspired of God as was Moses when he
received the Written Law on Sinai. Emanuel Deutsch is of the opinion
that from the very beginning of the Mosaic law there must have
existed a number of corollary laws which were used to interpret and
explain the written rules; that, besides, there were certain
enactments of the primitive Council of the Desert, and certain
verdicts issued by the later "judges within the gates"—all of which
entered into the general body of the Oral Law and were transmitted
side by side with the Written Law through the ages.[56] The fourth
book of Ezra, as well as other Apocryphal writings, together with
Philo and certain of the Church Fathers, tells us of great numbers of
books that were given to Moses at the same time that he received
the Pentateuch. These writings are doubtless the source of the
popular belief among the Jews that the traditional laws of the
Mishna had existed from time immemorial and were of divine origin.
"Jewish tradition traces the bulk of the oral injunctions, through a
chain of distinctly named authorities, to 'Sinai itself.' It mentions in
detail how Moses communicated those minutiæ of his legislation, in
which he had been instructed during the mysterious forty days and
nights on the Mount, to the chosen guides of the people, in such a
manner that they should forever remain engraven on the tablets of](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-63-320.jpg)
![their hearts."[57] This direct descent of the Oral Law from the Sacred
Mount itself would indicate an independent character and authority.
Nevertheless, Talmudic interpretation of tradition professed to
remain always subject to the Mosaic Code; to be built upon, and to
derive its highest inspiration from it. But, as a matter of fact, while
claiming theoretically to be subordinate to it, the Talmud finally
superseded and virtually displaced the Pentateuch as a legal and
administrative code. This was the inevitable consequence and effect
of the laws of growth and progress in national existence. Altered
conditions of life, at home and in exile, necessitated new rules of
action in the government of the Jewish commonwealth. The Mosaic
Code was found inadequate to the ever-changing exigencies of
Hebrew life. As a matter of fact, Moses laid down only general
principles for the guidance of Hebrew judges. He furnished the body
of the law, but a system of legal procedure was wholly wanting. The
Talmud supplied the deficiency and completed a perfect whole.
While yet in the Wilderness, Moses commanded the Israelites to
establish courts and appoint judges for the administration of justice
as soon as they were settled in Palestine.[58] This clearly indicates
that the great lawgiver did not intend his ordinances and injunctions
to be final and exclusive. Having furnished a foundation for the
scheme, he anticipated that the piety, judgment, and learning of
subsequent ages would do the rest. His expectations were fulfilled in
the development of the traditions afterwards embodied in the
Mishna, which is the principal component part of the Talmud.
As before suggested, with the growth in population and the ever-
increasing complications in social, political, and religious life, and
with the general advance in Hebrew civilization, Mosaic injunction
began to prove entirely inadequate to the national wants. In the
time intervening between the destruction of the first and second
Temples, a number of Mosaic laws had become utter anachronisms;
others were perfectly impracticable, and several were no longer even
understood. The exigencies of an altered mode of life and the
changed conditions and circumstances of the people rendered
imperative the enactment of new laws unknown to the Pentateuch.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-64-320.jpg)
![But the divine origin of the Hebrew system of law was never for a
moment forgotten, whatever the change and wherever made. The
Rabbins never formally repealed or abolished any Mosaic enactment.
They simply declared that it had fallen into desuetude. And, in
devising new laws rendered necessary by changed conditions of life
they invariably invoked some principle or interpretation of the
Written Law.
In the declining years of Jewish nationality, many characteristic laws
of the Pentateuch had become obsolete. The ordinance which
determined the punishment of a stubborn and rebellious son; the
enactment which commanded the destruction of a city given to
idolatry; and, above all, the lex talionis had become purely matters
of legend. On the other hand, many new laws appear in the Talmud
of which no trace whatever can be discovered in the Pentateuch.
"The Pharisees," says Josephus, "have imposed upon the people
many laws taken from the tradition of the Fathers, which are not
written in the law of Moses."[59] The most significant of these is the
one providing for Antecedent Warning in criminal prosecutions, the
meaning and purpose of which will be fully discussed in another
chapter.
Vicissitudes of the Talmud.—An old Latin adage runs: "Habent sua
fata libelli."[60] (Even books are victims of fate). This saying is
peculiarly applicable to the Talmud, which has had, in a general way,
the same fateful history as the race that created it. Proscription,
exile, imprisonment, confiscation, and burning was its lot throughout
the Middle Ages. During a thousand years, popes and kings vied
with each other in pronouncing edicts and hurling anathemas
against it. During the latter half of the sixteenth century it was
burned not fewer than six different times by royal or papal decree.
Whole wagonloads were consigned to the flames at a single burning.
In 1286, in a letter to the Archbishop of Canterbury, Honorius IV
described the Talmud as a "damnable book" (liber damnabilis), and
vehemently urged that nobody in England be permitted to read it,
since "all other evils flow out of it."[61] On New Year's day, 1553,](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-65-320.jpg)
![numerous copies of the Talmud were burned at Rome in compliance
with a decree of the Inquisition. And, as late as 1757, in Poland,
Bishop Dembowski, at the instigation of the Frankists, convened a
public assembly at Kamenetz-Podolsk, which decreed that all copies
of the Talmud found in the bishopric should be confiscated and
burned by the hangman.[62]
Of the two recensions, the Babylonian Talmud bore the brunt of
persecution during all the ages. This resulted from the fact that the
Jerusalem Talmud was little read after the closing of the Jewish
academies in Palestine, while the Babylonian Talmud was the
popular edition of eminent Jewish scholars throughout the world.
It is needless to say that the treatment accorded the venerable
literary compilation was due to bitter prejudice and crass ignorance.
This is well illustrated by the circumstance that when, in 1307,
Clement V was asked to issue a bull against the Talmud, he declined
to do so, until he had learned something about it. To his amazement
and chagrin, he could find no one who could throw any light upon
the subject. Those who wished it condemned and burned were
totally ignorant of its meaning and contents. The surprise and
disgust of Clement were so great that he resolved to found three
chairs in Hebrew, Arabic, and Chaldee, the three tongues nearest the
idiom of the Talmud. He designated the Universities of Paris,
Salamanca, Bologna, and Oxford as places where these languages
should be taught, and expressed the hope that, in time, one of these
universities might be able to produce a translation of "this
mysterious book."[63] It may be added that these plans of the Pope
were never consummated.
The Message and Mission of the Talmud.—To appreciate the
message and mission of the Talmud, its contents must be viewed
and contemplated in the light of both literature and history. As a
literary production it is a masterpiece—strange, weird, and unique—
but a masterpiece, nevertheless. It is a sort of spiritual and
intellectual cosmos in which the brain growth and soul burst of a
great race found expression during a thousand years. As an](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-66-320.jpg)
![CHAPTER II
HEBREW CRIMINAL LAW—CRIMES AND
PUNISHMENTS
apital crimes, under Hebrew law, were classified by
Maimonides according to their respective penalties. His
arrangement will be followed in this chapter.[64]
Hebrew jurisprudence provided four methods of capital
punishment: (1) Beheading; (2) Strangling; (3) Burning;
(4) Stoning.
Crucifixion was unknown to Hebrew law. This cruel and loathsome
form of punishment will be fully discussed in the second volume of
this work.
Thirty-six capital crimes are mentioned by the Pentateuch and the
Talmud.
Beheading was the punishment for only two crimes:
(1) Murder.
(2) Communal apostasy from Judaism to idolatry.
Strangling was prescribed for six offenses:
(1) Adultery.
(2) Kidnaping.
(3) False prophecy.
(4) Bruising a parent.
(5) Prophesying in the name of heathen deities.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-69-320.jpg)
![(6) Maladministration (the "Rebellious Elder").
Burning was the death penalty for ten forms of incest—criminal
commerce:
(1) With one's own daughter.
(2) With one's own son's daughter.
(3) With one's own daughter's daughter.
(4) With one's own stepdaughter.
(5) With one's own stepson's daughter.
(6) With one's own stepdaughter's daughter.
(7) With one's own mother-in-law.
(8) With one's own mother-in-law's mother.
(9) With one's own father-in-law's mother.
(10) With a priest's daughter.[65]
Stoning was the penalty for eighteen capital offenses:
(1) Magic.
(2) Idolatry.
(3) Blasphemy.
(4) Pythonism.
(5) Pederasty.
(6) Necromancy.
(7) Cursing a parent.
(8) Violating the Sabbath.
(9) Bestiality, practiced by a man.
(10) Bestiality, practiced by a woman.
(11) Sacrificing one's own children to Moloch.
(12) Instigating individuals to embrace idolatry.
(13) Instigating communities to embrace idolatry.
(14) Criminal conversation with one's own mother.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-70-320.jpg)
![(15) Criminal conversation with a betrothed virgin.
(16) Criminal conversation with one's own stepmother.
(17) Criminal conversation with one's own daughter-in-law.
(18) Violation of filial duty (making the "Prodigal Son").[66]
The crime of false swearing requires special notice. This offense
could not be classified under any of the above subdivisions because
of its peculiar nature. The Mosaic Code ordains in Deut. xix. 16-21:
"If a false witness rise up against any man to testify against him that
which is wrong ... and, behold, if the witness be a false witness, and
hath testified falsely against his brother; then shall ye do unto him,
as he had thought to have done unto his brother ... and thine eye
shall not pity, but life shall go for life, eye for eye, tooth for tooth,
hand for hand, foot for foot." Talmudic construction of this law
awarded the same kind of death to him who had sworn falsely
against his brother that would have been meted out to the alleged
criminal, if the testimony of the false swearer had been true.
Imprisonment, as a method of punishment, was unknown to the
Mosaic Code. Leviticus xxiv. 12 and Numbers xv. 34 seem to indicate
the contrary; but the imprisonment therein mentioned undoubtedly
refers to the mere detention of the prisoner until sentence could be
pronounced against him. Imprisonment as a form of punishment
was a creation of the Talmudists who legalized its application among
the Hebrews. According to Mendelsohn, five different classes of
offenders were punished by imprisonment:
(1) Homicides; whose crime could not be legally punished with
death, because some condition or other, necessary to produce a
legal conviction, had not been complied with.
(2) Instigators to or procurers of murder; such, for instance, as had
the deed committed by the hands of a hireling.
(3) Accessories to loss of life, as, for instance, when several persons
had clubbed one to death, and the court could not determine the
one who gave the death blow.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-71-320.jpg)
![(4) Persons who having been twice duly condemned to and punished
with flagellation for as many transgressions of one and the same
negative precept, committed it a third time.
(5) Incorrigible offenders, who, on each of three occasions, had
failed to acknowledge as many warnings antecedent to the
commission of one and the same crime, the original penalty for
which was excision.[67]
Flagellation is the only corporal punishment mentioned by the
Pentateuch. The number of stripes administered were not to exceed
forty and were to be imposed in the presence of the judges.[68]
Wherever the Mosaic Code forbade an act, or, in the language of the
sages, said "Thou shalt not," and prescribed no other punishment or
alternative, a Court of Three might impose stripes as the penalty for
wrongdoing. Mendelsohn gives the following classification:
Flagellation is the penalty of three classes of offenses:
(1) The violation of a negative precept, deadly in the sight of
heaven.
(2) The violation of any negative precept, when accomplished by
means of a positive act.
(3) The violation of any one of the prohibitive ordinances punishable,
according to the Mosaic law with excision, to which, however, no
capital punishment at the instance of a human tribunal is attached.
[69]
The Mishna enumerates fifty offenses punishable by stripes, but this
enumeration is evidently incomplete. Maimonides gives a full
classification of all the offenses punishable by flagellation, the
number of which he estimates to be two hundred and seven. The
last three in his list are cases in which the king takes too many
wives, accumulates too much silver or gold, or collects too many
horses.[70]](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-72-320.jpg)
![unfortunate and downtrodden of the earth; debtors, slaves,
criminals, and political offenders; some sacred spot—an altar, a
grave, or a sanctuary dedicated and devoted to some divinity who
threw about the hallowed place divine protection and inviolability.
Such was at Athens the Temple of Theseus, the sanctuary of slaves.
It will be remembered that the orator Demosthenes took refuge in
the Temple of Poseidon as a sanctuary, when pursued by emissaries
of Antipater and the Macedonians.[71] Among the ancient Hebrews,
there were six cities of refuge; three on either side of the Jordan.
They were so located as to be nearly opposite each other. Bezer in
Reuben was opposite Hebron in Judah; Schechem in Ephraim was
opposite to Ramoth in Gad; and Golan in Manasseh was opposite to
Kedesh in Naphtali.[72] Highways in excellent condition led from one
to the other. Signposts were placed at regular intervals to indicate
the way to the nearest city of refuge. These cities were designated
by the law as asylums or sanctuaries for the protection of innocent
slayers of their fellow-men from the "avenger of blood." Among
nearly all primitive peoples of crude political development, such as
the early Germans, the ancient Greeks and Slavs, certain North
American savage tribes and the modern Arabs, Corsicans and
Sicilians, the right of private vengeance was and is taught and
tolerated. Upon the "next of kin," the "avenger of blood," devolved
the duty of hunting down and slaying the guilty man. Cities of refuge
were provided by Mosaic law for such an emergency among the
Hebrews. This provision of the Mosaic Code doubtless sprang from a
personal experience of its founder. Bible students will remember that
Moses slew an Egyptian and was compelled to flee in consequence.
[73] Remembering his dire distress on this occasion, the great
lawgiver was naturally disposed to provide sanctuaries for others
similarly distressed. But the popular notion of the rights of sanctuary
under the Mosaic law is far from right. That a common murderer
could, by precipitate flight, reach one of the designated places and
be safe from his pursuers and the vengeance of the law, is thought
by many. The observation of Benny on this point is apt and lucid:](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-74-320.jpg)
![Internment in one of the cities of refuge was not the
scampering process depicted in the popular engraving: a man
in the last stage of exhaustion at the gate of an Eastern
town; his pursuers close upon him, arrows fixed and bows
drawn; his arms stretched imploringly towards a fair Jewish
damsel, with a pitcher gracefully poised upon her head. This
may be extremely picturesque, but it is miserably unlike the
custom in vogue among the later Hebrews. Internment in a
city of refuge was a sober and judicial proceeding. He who
claimed the privilege was tried before the Sanhedrin like any
ordinary criminal. He was required to undergo examination;
to confront witnesses, to produce evidence, precisely as in
the case of other offenders. He had to prove that the
homicide was purely accidental; that he had borne no malice
against his neighbor; that he had not lain in wait for him to
slay him. Only when the judges were convinced that the
crime was homicide by misadventure was the culprit
adjudged to be interned in one of the sheltering cities. There
was no scurrying in the matter; no abrupt flight; no hot
pursuit, and no appeal for shelter. As soon as judgment was
pronounced the criminal was conducted to one of the
appointed places. He was accompanied the whole distance by
two talmide-chachamin-disciples of the Rabbins. The
avengers of the blood dared not interfere with the offender
on the way. To slay him would have been murder, punishable
with death.
Execution of Capital Sentences. (1) Beheading.—The Hebrews
considered beheading the most awful and ignominious of all forms of
punishment. It was the penalty for deliberate murder and for
communal apostasy from Judaism to idolatry, the most heinous
offenses against the Hebrew theocracy. Beheading was accomplished
by fastening the culprit securely to a post and then severing his
head from his body by a stroke with a sword.[74]](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-75-320.jpg)
![(2) Strangling.—The capital punishment of strangling was effected
by burying the culprit to his waist in soft mud, and then tightening a
cord wrapped in a soft cloth around his neck, until suffocation
ensued.[75]
(3) Burning.—The execution of criminals by burning was not done by
consuming the living person with fire, as was practiced in the case of
heretics by prelates in the Middle Ages and in the case of white
captives by savages in colonial days in America. Indeed, the term
"burning" seems to be a misnomer in this connection, for the culprit
was not really burned to death. He was simply suffocated by
strangling. As in the case of strangling, the condemned man was
placed in a pit dug in the ground. Soft dirt was then thrown in and
battered down, until nothing but his head and chest protruded. A
cord, wrapped in a soft cloth, was then passed once around his
neck. Two strong men came forward, grasped each an end, and
drew the cord so hard that suffocation immediately followed. As the
lower jaw dropped from insensibility and relaxation, a lighted wick
was quickly thrown into his mouth. This constituted the burning.[76]
There is authority for the statement that instead of a lighted wick,
molten lead was poured down the culprit's throat.[77]
(4) Stoning.—Death by stoning was accomplished in the following
manner: The culprit was taken to some lofty hill or eminence, made
to undress completely, if a man, and was then precipitated violently
to the ground beneath. The fall usually broke the neck or dislocated
the spinal cord. If death did not follow instantaneously the witnesses
hurled upon his prostrate body heavy stones until he was dead. If
the first stone, so heavy as to require two persons to carry it, did not
produce death, then bystanders threw stones upon him until death
ensued. Here, again, "stoning" to death is not strictly accurate.
Death usually resulted from the fall of the man from the platform,
scaffold, hill, or other elevation from which he was hurled. It was
really a process of neck-breaking, instead of stoning, as burning was
a process of suffocation, instead of consuming with fire.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-76-320.jpg)
![CHAPTER III
HEBREW CRIMINAL LAW—COURTS AND
JUDGES
HE Hebrew tribunals were three in kind: the Great
Sanhedrin; the Minor Sanhedrin; and the Lower
Tribunal, or the Court of Three.
The Great Sanhedrin, or Grand Council, was the high
court of justice and the supreme tribunal of the Jews. It
sat at Jerusalem. It numbered seventy-one members. Its powers
were legislative, executive, and judicial. It exercised all the functions
of education, of government, and of religion. It was the national
parliament of the Hebrew Theocracy, the human administrator of the
divine will. It was the most august tribunal that ever interpreted or
administered religion to man.
The Name.—The word "Sanhedrin" is derived from the Greek
(συνέδριον) and denotes a legislative assembly or an ecclesiastical
council deliberating in a sitting posture. It suggests also the gravity
and solemnity of an Oriental synod, transacting business of great
importance. The etymology of the word indicates that it was first
used in the later years of Jewish nationality. Several other names are
also found in history to designate the Great Sanhedrin of the Jews.
The Council of Ancients is a familiar designation of early Jewish
writers. It is called Gerusia, or Senate, in the second book of
Maccabees.[78] Concilium, or Grand Council, is the name found in
the Vulgate.[79] The Talmud designates it sometimes as the Tribunal
of the Maccabees, but usually terms it Sanhedrin, the name most
frequently employed in the Greek text of the Gospels, in the writings
of the Rabbins, and in the works of Josephus.[80]](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-78-320.jpg)
![Origin of the Great Sanhedrin.—The historians are at loggerheads as
to the origin of the Great Sanhedrin. Many contend that it was
established in the Wilderness by Moses, who acted under divine
commission recorded in Numbers xi. 16, 17: "Gather unto me
seventy of the elders of Israel, whom thou knowest to be the elders
of the people, and officers of them; and bring them unto the
tabernacle of the congregation, that they may stand with thee; and I
will take of the Spirit that is upon thee and will put it upon them;
and they shall bear the burden of the people with thee, that thou
bearest it not alone." Over the seventy elders, Moses is said to have
presided, making seventy-one, the historic number of the Great
Sanhedrin. Several Christian historians, among them Grotius and
Selden, have entertained this view; others equally celebrated have
maintained contrary opinions. These latter contend that the council
of seventy ordained by Moses existed only a short time, having been
established to assist the great lawgiver in the administration of
justice; and that, upon the entrance of the children of Israel into the
Promised Land, it disappeared altogether. The writers who hold this
view contend that if the great assembly organized in the Wilderness
was perpetuated side by side with the royal power, throughout the
ages, as the Rabbis maintained, some mention of this fact would, in
reason, have been made by the Bible, Josephus, or Philo.
The pages of Jewish history disclose the greatest diversity of opinion
as to the origin of the Great Sanhedrin. The Maccabean era is
thought by some to be the time of its first appearance. Others
contend that the reign of John Hyrcanus, and still others that the
days of Judas Maccabeus, marked its birth and beginning. Raphall,
having studied with care its origin and progress, wrote: "We have
thus traced the existence of a council of Zekenim or Elders founded
by Moses, existing in the days of Ezekiel, restored under the name of
Sabay Yehoudai, or Elders of the Jews, under Persian dominion;
Gerusia, under the supremacy of the Greeks; and Sanhedrin under
the Asmonean kings and under the Romans."[81]](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-79-320.jpg)
![Brushing aside mere theory and speculation, one historical fact is
clear and uncontradicted, that the first Sanhedrin Council clothed
with the general judicial and religious attributes of the Great
Sanhedrin of the times of Jesus, was established at Jerusalem
between 170 and 106 B.C.
Organization of the Great Sanhedrin.—The seventy-one members
composing the Great Sanhedrin were divided into three chambers:
The chamber of priests;
The chamber of scribes;
The chamber of elders.
The first of these orders represented the religious or sacerdotal; the
second, the literary or legal; the third, the patriarchal, the
democratic or popular element of the Hebrew population. Thus the
principal Estates of the Commonwealth of Israel were present, by
representation, in the great court and parliament of the nation.
Matthew refers to these three orders and identifies the tribunal that
passed judgment upon Christ: "From that time forth, began Jesus to
shew unto his disciples, how that he must go unto Jerusalem, and
suffer many things of the elders and chief priests and scribes, and
be killed and raised again the third day."[82]
Theoretically, under the Hebrew constitution, the "seventy-one" of
the three chambers were to be equally divided:
Twenty-three in the chamber of priests,
Twenty-three in the chamber of scribes,
Twenty-three in the chamber of elders.
A total of sixty-nine, together with the two presiding officers, would
constitute the requisite number, seventy-one. But, practically, this
arrangement was rarely ever observed. The theocratic structure of
the government of Israel and the pious regard of the people for the](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-80-320.jpg)
![guardians of the Temple, gave the priestly element a predominating
influence from time to time. The scribes, too, were a most vigorous
and aggressive sect and frequently encroached upon the rights and
privileges of the other orders. Abarbanel, one of the greatest of the
Hebrew writers, has offered this explanation: "The priests and
scribes naturally predominated in the Sanhedrin because, not having
like the other Israelites received lands to cultivate and improve, they
had abundant time to consecrate to the study of law and justice, and
thus became better qualified to act as judges."[83]
Qualifications of Members of the Great Sanhedrin.—The following
qualifications were requisite to entitle an applicant to membership in
the Great Sanhedrin:
(1) He must have been a Hebrew and a lineal descendant of Hebrew
parents.[84]
(2) He must have been "learned in the law"; both written and
unwritten.
His legal attainment must have included an intimate acquaintance
with all the enactments of the Mosaic Code, with traditional
practices, with the precepts and precedents of the colleges, with the
adjudications of former courts and the opinions of former judges. He
must have been familiar not only with the laws then actively in force,
but also with those that had become obsolete.[85]
(3) He must have had judicial experience; that is, he must have
already filled three offices of gradually increasing dignity, beginning
with one of the local courts, and passing successively through two
magistracies at Jerusalem.[86]
(4) He must have been thoroughly proficient in scientific knowledge.
The ancient Sanhedrists were required to be especially well
grounded in astronomy and medicine. They were also expected to
be familiar with the arts of the necromancer.[87] We are also led to
believe from the revelations of the Talmud that the judges of Israel](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-81-320.jpg)
![were well versed in the principles of physiology and chemistry, as far
as these sciences were developed and understood in those days.
History records that Rabbi Ismael and his disciples once engaged in
experimental dissection in order to learn the anatomy of the human
frame. On one occasion a deceitful witness tried to impose upon a
Hebrew court by representing spermatic fluid to be the albumen of
an egg. Baba bar Boutah was enabled, from his knowledge of the
elements of chemistry, to demonstrate the fact of fraud in the
testimony of the witness. Eighty disciples of the famous Academy of
Hillel are said to have been acquainted with every branch of science
known in those days.[88]
(5) He must have been an accomplished linguist; that is, he must
have been thoroughly familiar with the languages of the surrounding
nations.
Interpreters were not allowed in Hebrew courts. A knowledge of
several languages was, therefore, indispensable to the candidate
who sought membership in the Great Sanhedrin. "In the case of a
foreigner being called as a witness before a tribunal, it was
absolutely necessary that two members should understand the
language in which the stranger's evidence was given; that two
others should speak to him; while another was required to be both
able to understand and to converse with the witness. A majority of
three judges could always be obtained on any doubtful point in the
interpretation of the testimony submitted to the court. At Bither
there were three Rabbins acquainted with every language then
known, while at Jabneh there were said to be four similarly endowed
with the gift of 'all the tongues.'"[89]
(6) He must have been modest, popular, of good appearance, and
free from haughtiness.[90]
The Hebrew mind conceived modesty to be the natural result of that
learning, dignity, and piety which every judge was supposed to
possess. The qualification of "popularity" did not convey the notion
of electioneering, hobnobbing and familiarity. It meant simply that](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-82-320.jpg)
![the reputation of the applicant for judicial honors was so far above
reproach that his countrymen could and would willingly commit all
their interests of life, liberty, and property to his keeping. By "good
appearance" was meant that freedom from physical blemishes and
defects, and that possession of physical endowments that would
inspire respect and reverence in the beholder. The haughty judge
was supposed to be lacking in the elements of piety and humility
which qualified him for communion with God. Haughtiness,
therefore, disqualified for admission to the Great Sanhedrin.
(7) He must have been pious, strong, and courageous.[91]
Piety was the preëminent qualification of a judge of Israel. Impiety
was the negation of everything Israelitish. Strength and courage are
attributes that all judges in all ages and among all races have been
supposed to possess in order to be just and righteous in their
judgments.
Disqualifications.—Disqualifications of applicants for membership in
the Great Sanhedrin are not less interesting than qualifications. They
are in the main mere negatives of affirmatives which have already
been given, and would seem, therefore, to be superfluous. But they
are strongly accentuated in Hebrew law, and are therefore repeated
here.
(1) A man was disqualified to act as judge who had not, or had
never had, any regular trade, occupation, or profession by which he
gained his livelihood.
The reason for this disqualification was based upon a stringent
maxim of the Rabbins: "He who neglects to teach his son a trade, is
as though he taught him to steal!" A man who did not work and had
never labored in the sweat of his brow for an honest livelihood, was
not qualified, reasoned the Hebrew people, to give proper
consideration or extend due sympathy to the cause of litigants
whose differences arose out of the struggles of everyday life.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-83-320.jpg)
![(2) In trials where the death penalty might be inflicted, an aged
man, a person who had never had any children of his own, and a
bastard were disqualified to act as judge.
A person of advanced years was disqualified because according to
the Rabbins old age is frequently marked by bad temper; and
"because his years and infirmities were likely to render him harsh,
perhaps obstinate and unyielding." On the other hand, youth was
also a disqualification to sit in the Sanhedrin. According to the
Rabbis, twenty-five years was the age which entitled a person to be
called a Man;[92] but no one was eligible to a seat in the Sanhedrin
until he had reached the age of forty years.[93] The ancient Hebrews
regarded that period as the beginning of discretion and
understanding.
A person without children was not supposed to possess those tender
paternal feelings "which should warm him on behalf of the son of
Israel who was in peril of his life."
The stain of birth and the degradation in character of a bastard were
wholly inconsistent with the high ideals of the qualifications of a
Hebrew judge.
(3) Gamblers, dice players, bettors on pigeon matches, usurers, and
slave dealers were disqualified to act as judges.
The Hebrews regarded gambling, dice playing, betting on pigeon
matches, and other such practices as forms of thievery; and thieves
were not eligible to sit as judges in their courts. No man who was in
the habit of lending money in an usurious manner could be a judge.
It was immaterial whether the money was lent to a countryman or a
stranger. Slave dealers were disqualified to act as judges because
they were regarded as inhuman and unsympathetic.
(4) No man was qualified to be a judge who had dealt in the fruits of
the seventh year.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-84-320.jpg)
![(nasi), was the chief and the president of the court. The other,
known as the father of the Tribunal (ab-beth-din), was the vice-
president.
There has been much discussion among the historians as to the
particular chamber from which the president was chosen. Some have
contended that the presidency of the Sanhedrin belonged by right to
the high priest. But the facts of history do not sustain this
contention. Aaron was high priest at the time when Moses was
president of the first Sanhedrin in the Wilderness; and, besides, the
list of presidents preserved by the Talmud reveals the names of
many who did not belong to the priesthood. Maimonides has made
the following very apt observation on the subject: "Whoever
surpassed his colleagues in wisdom was made by them chief of the
Sanhedrin."[94]
According to most Jewish writers, there were two scribes or
secretaries of the Sanhedrin. But several others contend that there
were three. Benny says: "Three scribes were present; one was
seated on the right, one on the left, the third in the center of the
hall. The first recorded the names of the judges who voted for the
acquittal of the accused, and the arguments upon which the
acquittal was grounded. The second noted the names of such as
decided to condemn the prisoner and the reasons upon which the
conviction was based. The third kept an account of both the
preceding so as to be able at any time to supply omissions or check
inaccuracies in the memoranda of his brother reporters."[95]
In addition to these officers, there were still others who executed
sentences and attended to all the police work of legal procedure.
They were called shoterim.[96]
There was no such officer as a public prosecutor or State's attorney
known to the laws of the ancient Hebrews. The witnesses to the
crime were the only prosecutors recognized by Hebrew criminal
jurisprudence; and in capital cases they were the legal executioners
as well.](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-86-320.jpg)
![There was also no such body as the modern Grand Jury known to
ancient Hebrew criminal law. And no similar body of committee of
the Sanhedrin performed the accusatory functions of the modern
Grand Jury. The witnesses were the only accusers, and their
testimony was both the indictment and the evidence. Until they
testified, the man suspected was deemed not only innocent but
unaccused.
The profession of the law, in the modern sense of the term, was no
part of the judicial system of the ancient Hebrews. There were no
advocates as we know them. There were, indeed, men learned in
the law—Pharisees and Sadducees—who knew all the law. There
were doctors of the law: men whom Jesus confounded when a youth
in the Temple at the age of twelve.[97] But there were no lawyers in
the modern sense: professional characters who accept fees and
prosecute cases. The judges and disciples performed all the duties of
the modern attorney and counselor-at-law. The prophets were the
sole orators of Hebrew life, but they were never allowed to appear
as defendants of accused persons. Indeed, they themselves were at
times compelled to play the role of defendants. Jeremiah is an
illustrious example.[98] Both Keim[99] and Geikie[100] speak of a Baal
Rib, a counsel appointed to see that everything possible was done to
secure the rights of an accused person at a Hebrew criminal trial.
But these statements are not in accord with standard works on
ancient Hebrew jurisprudence. Indeed, Friedlieb emphatically denies
that there was any such person as a Baal Rib or Dominus Litis
among the ancient Hebrews.[101] It seems that in the closing years
of Jewish nationality, specially retained advocates were known, for
St. Luke tells us that the Jews employed Tertullus, a certain orator,
to prosecute St. Paul.[102] But this was certainly an exceptional case.
It is historically certain that in the early ages of the Jewish
Commonwealth litigants pleaded their own causes. This we learn
from the case of the two women who appeared before King
Solomon, and laid before him their respective claims to a child.[103]](https://image.slidesharecdn.com/2611904-250620134036-7512c432/85/Yeast-Functional-Genomics-Methods-And-Protocols-Frdric-Devaux-87-320.jpg)
Yeast Functional Genomics Methods And Protocols Frdric Devaux
- 1. Yeast Functional Genomics Methods And Protocols Frdric Devaux download https://ebookbell.com/product/yeast-functional-genomics-methods- and-protocols-frdric-devaux-5223808 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Yeast Functional Genomics And Proteomics Methods And Protocols 1st Edition Barry Dion https://ebookbell.com/product/yeast-functional-genomics-and- proteomics-methods-and-protocols-1st-edition-barry-dion-1647346 Investigations In Yeast Functional Genomics And Molecular Biology 1st Edition Matthew Eckwahl https://ebookbell.com/product/investigations-in-yeast-functional- genomics-and-molecular-biology-1st-edition-matthew-eckwahl-4748012 Guide To Yeast Genetics Functional Genomics Proteomics And Other Systems Analysis 2nd Edition Jonathan Weissman https://ebookbell.com/product/guide-to-yeast-genetics-functional- genomics-proteomics-and-other-systems-analysis-2nd-edition-jonathan- weissman-1479836 Functional Genetics Of Industrial Yeasts 1st Edition Han De Winde Auth https://ebookbell.com/product/functional-genetics-of-industrial- yeasts-1st-edition-han-de-winde-auth-4326276
- 3. Structure Function And Regulation Of Tor Complexes From Yeasts To Mammals Part A 1st Edition Michael N Hall https://ebookbell.com/product/structure-function-and-regulation-of- tor-complexes-from-yeasts-to-mammals-part-a-1st-edition-michael-n- hall-2375672 Structure Function And Regulation Of Tor Complexes From Yeasts To Mammals Part B 1st Edition Fuyuhiko Tamanoi https://ebookbell.com/product/structure-function-and-regulation-of- tor-complexes-from-yeasts-to-mammals-part-b-1st-edition-fuyuhiko- tamanoi-2435458 Yeast Surface Display 1st Michael W Traxlmayr https://ebookbell.com/product/yeast-surface-display-1st-michael-w- traxlmayr-47937320 Yeast The Practical Guide To Beer Fermentation Brewing Elements Illustrated White https://ebookbell.com/product/yeast-the-practical-guide-to-beer- fermentation-brewing-elements-illustrated-white-54873930 Yeast As A Tool In Cancer Research 1st Edition John L Nitiss Editor https://ebookbell.com/product/yeast-as-a-tool-in-cancer-research-1st- edition-john-l-nitiss-editor-2020804
- 5. Yeast Functional Genomics Frédéric Devaux Editor Methods and Protocols Methods in Molecular Biology 1361
- 6. ME T H O D S I N MO L E C U L A R BI O L O G Y Series Editor John M. Walker School of Life and Medical Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK For further volumes: http://www.springer.com/series/7651
- 8. Yeast Functional Genomics Methods and Protocols Edited by Frédéric Devaux Laboratoire de biologie computationnelle et quantitative,Sorbonne Universités, Paris,France
- 9. ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-4939-3078-4 ISBN 978-1-4939-3079-1 (eBook) DOI 10.1007/978-1-4939-3079-1 Library of Congress Control Number: 2015951231 Springer New York Heidelberg Dordrecht London © Springer Science+Business Media New York 2016 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. Printed on acid-free paper Humana Press is a brand of Springer Springer Science+Business Media LLC New York is part of Springer Science+Business Media (www.springer.com) Editor Frédéric Devaux Laboratoire de biologie computationnelle et quantitative Sorbonne Universités Paris, France
- 10. v This volume of the “Methods in Molecular Biology” series aims at reflecting the state of the art of yeast functional genomics. Since the publication of its genome sequence in 1996, yeast functional genomics has been at the forefront of technological advances and never stopped evolving. Ten years ago, 90 % of the publications in this field were made of micro- array-based transcriptome and chromatin immunoprecipitation analyses, and the reader will find in this volume the most recent protocols for these “classics” which are still widely used and up to date. Since then, yeast functional genomics have diversified in many ways. First, the emergence of high-throughput sequencing technologies considerably enlarged our capacity to investigate yeast transcriptomes and genomes. Hence most of the chapters of this volume present protocols based on new generation sequencing technologies. Second, all aspects of gene expression regulation, from nuclear architecture to transla- tional rates and metabolite steady states, can now be studied at a genome-wide scale. This volume provides a panel of protocols for the study of DNA-DNA contact maps, replication profiles, transcription rates, RNA secondary structures, protein-RNA interactions, ribo- some profiling, and quantitative proteomes and metabolomes. Third, the availability of genome sequences for tens of yeast species and hundreds of strains in some species allowed for yeast comparative functional genomics and yeast popula- tions genomics and opened the way to a common use of the natural or laboratory-gener- ated genetic polymorphism to identify functional relationships between genes and gene-phenotype interactions in a powerful and comprehensive way. This volume includes protocols for yeast comparative transcriptomics, yeast high-throughput genetic screens, yeast QTL mapping, and yeast experimental evolution. Moreover, several protocols pre- sented here were optimized for other species than S. cerevisiae. Finally, the accumulation of these genome-wide data of various natures pushed forward the development of bioinformatics tools and methods to make available, represent, and analyze the properties of large yeast cellular networks. Most of the protocols presented in this volume emphasized both “wet lab” and in silico analyses aspects. Moreover, two chap- ters were specifically dedicated to the integration of high-throughput data in evolutionary models and to data mining of global regulatory networks, respectively. Obviously, the field is so diverse that this book could not be comprehensive. For instance, just the different methods nowadays available for yeast quantitative proteomics would have filled the whole volume. Our goal was rather to make this issue of Methods in Molecular Biology as representative of its time and as useful to a broad audience as possible. Did we achieve this goal? I believe the answer is yes but actually, this is to the reader to tell. So… Have a nice reading! Paris, France Frédéric Devaux Preface
- 12. vii Contents Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix 1 Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Can Wang, Markus S. Schröder, Stephen Hammel, and Geraldine Butler 2 Enhancing Structural Annotation of Yeast Genomes with RNA-Seq Data . . . . 41 Hugo Devillers, Nicolas Morin, and Cécile Neuvéglise 3 Pathogen Gene Expression Profiling During Infection Using a Nanostring nCounter Platform. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Wenjie Xu, Norma V. Solis, Scott G. Filler, and Aaron P. Mitchell 4 Comparative Transcriptomics in Yeasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Dawn A. Thompson 5 Mapping the Transcriptome-Wide Landscape of RBP Binding Sites Using gPAR-CLIP-seq: Experimental Procedures . . . . . . . . . . . . . . . . . . . . . . 77 Ting Han and John K. Kim 6 Mapping the Transcriptome-Wide Landscape of RBP Binding Sites Using gPAR-CLIP-seq: Bioinformatic Analysis . . . . . . . . . . . . . . . . . . . . . . . . 91 Mallory A. Freeberg and John K. Kim 7 Translation Analysis at the Genome Scale by Ribosome Profiling. . . . . . . . . . . 105 Agnès Baudin-Baillieu, Isabelle Hatin, Rachel Legendre, and Olivier Namy 8 Biotin-Genomic Run-On (Bio-GRO): A High-Resolution Method for the Analysis of Nascent Transcription in Yeast . . . . . . . . . . . . . . . . . . . . . . 125 Antonio Jordán-Pla, Ana Miguel, Eva Serna, Vicent Pelechano, and José E. Pérez-Ortín 9 Genome-Wide Probing of RNA Structures In Vitro Using Nucleases and Deep Sequencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Yue Wan, Kun Qu, Zhengqing Ouyang, and Howard Y. Chang 10 Genome-Wide Chromatin Immunoprecipitation in Candida albicans and Other Yeasts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Matthew B. Lohse, Pisiwat Kongsomboonvech, Maria Madrigal, Aaron D. Hernday, and Clarissa J. Nobile 11 ChIPseq in Yeast Species: From Chromatin Immunoprecipitation to High-Throughput Sequencing and Bioinformatics Data Analyses . . . . . . . . 185 Gaëlle Lelandais, Corinne Blugeon, and Jawad Merhej 12 Systematic Determination of Transcription Factor DNA-Binding Specificities in Yeast. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Lourdes Peña-Castillo and Gwenael Badis
- 13. viii 13 Generation and Analysis of Chromosomal Contact Maps of Yeast Species . . . . 227 Axel Cournac, Martial Marbouty, Julien Mozziconacci, and Romain Koszul 14 A Versatile Procedure to Generate Genome-Wide Spatiotemporal Program of Replication in Yeast Species. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Nicolas Agier and Gilles Fischer 15 Single-Step Affinity Purification (ssAP) and Mass Spectrometry of Macromolecular Complexes in the Yeast S. cerevisiae . . . . . . . . . . . . . . . . . . 265 Christian Trahan, Lisbeth-Carolina Aguilar, and Marlene Oeffinger 16 Label-Free Quantitative Proteomics in Yeast . . . . . . . . . . . . . . . . . . . . . . . . . . 289 Thibaut Léger, Camille Garcia, Mathieu Videlier, and Jean-Michel Camadro 17 Profiling of Yeast Lipids by Shotgun Lipidomics . . . . . . . . . . . . . . . . . . . . . . . 309 Christian Klose and Kirill Tarasov 18 Identification of Links Between Cellular Pathways by Genetic Interaction Mapping (GIM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 Christophe Malabat and Cosmin Saveanu 19 On the Mapping of Epistatic Genetic Interactions in Natural Isolates: Combining Classical Genetics and Genomics. . . . . . . . . . . . . . . . . . . . . . . . . . 345 Jing Hou and Joseph Schacherer 20 Experimental Evolution and Resequencing Analysis of Yeast . . . . . . . . . . . . . . 361 Celia Payen and Maitreya J. Dunham 21 Reconstruction and Analysis of the Evolution of Modular Transcriptional Regulatory Programs Using Arboretum . . . . . . . . . . . . . . . . . 375 Sara A. Knaack, Dawn A. Thompson, and Sushmita Roy 22 Predicting Gene and Genomic Regulation in Saccharomyces cerevisiae, using the YEASTRACT Database: A Step-by-Step Guided Analysis . . . . . . . . . 391 Miguel C. Teixeira, Pedro T. Monteiro, and Isabel Sá-Correia Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 Contents
- 14. ix NICOLAS AGIER • Biologie Computationnelle et Quantitative, Sorbonne Universités, UPMC Univ. Paris 06, CNRS, UMR 7238, Paris, France; Biologie Computationnelle et Quantitative, CNRS-Université Pierre et Marie Curie, UMR 7238, Paris, France LISBETH-CAROLINA AGUILAR • Institut de recherches cliniques de Montréal, Montréal, QC, Canada GWENAEL BADIS • Institut Pasteur, Génétique des Interactions Macromoléculaires, Centre National de la Recherche Scientifique, Paris, France AGNÈS BAUDIN-BAILLIEU • Institute for Integrative Biology of the Cell (I2BC), UMR 9198 CEA, CNRS, Université Paris Sud, Orsay, France CORINNE BLUGEON • Plateforme Génomique, Ecole Normale Supérieure, Institut de Biologie de l’ENS, IBENS, Paris, France; Inserm, U1024, Paris, France; CNRS, UMR 8197, Paris, France GERALDINE BUTLER • School of Biomolecular and Biomedical Science, Conway Institute, University College Dublin, Belfield, Dublin, Ireland JEAN-MICHEL CAMADRO • Mass Spectrometry Laboratory, Institut Jacques Monod, UMR7592, CNRS—Univ Paris Diderot, Sorbonne Paris Cité, Paris, France; Mitochondria, Metals and Oxidative Stress group, Institut Jacques Monod, UMR7592, CNRS—Univ Paris Diderot, Sorbonne Paris Cité, Paris, France HOWARD Y. CHANG • Howard Hughes Medical Institute and Program in Epithelial Biology, Stanford University School of Medicine, Stanford, CA, USA AXEL COURNAC • Institut Pasteur, Department Genomes and Genetics, Groupe Régulation Spatiale des Génomes, Paris, France; CNRS, UMR 3525, Paris, France HUGO DEVILLERS • INRA, UMR1319 Micalis, Jouy-en-Josas, France; AgroParisTech, UMR Micalis, Jouy-en-Josas, France MAITREYA J. DUNHAM • Department of Genome Sciences, University of Washington, Seattle, WA, USA SCOTT G. FILLER • Division of Infectious Diseases, Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA, USA GILLES FISCHER • Biologie Computationnelle et Quantitative, Sorbonne Universités, UPMC Univ Paris 06, CNRS, UMR 7238, Paris, France; Biologie Computationnelle et Quantitative, CNRS-Université Pierre et Marie Curie, UMR 7238, Paris, France MALLORY A. FREEBERG • Life Sciences Institute, University of Michigan, Ann Arbor, MI, USA; Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI, USA CAMILLE GARCIA • Mass Spectrometry Laboratory, Institut Jacques Monod, UMR7592, CNRS—Univ Paris Diderot, Sorbonne Paris Cité, Paris Cedex 13, France STEPHEN HAMMEL • School of Biomolecular and Biomedical Science, Conway Institute, University College Dublin, Belfield, Dublin, Ireland TING HAN • Department of Biochemistry, UT Southwestern Medical Center, Dallas, TX, USA ISABELLE HATIN • Institute for Integrative Biology of the Cell (I2BC), UMR 9198 CEA, CNRS, Université Paris Sud, Orsay, France Contributors
- 15. x AARON D. HERNDAY • Department of Molecular and Cell Biology, School of Natural Sciences, University of California, Merced, Merced, CA, USA JING HOU • Department of Genetics, Genomics and Microbiology, CNRS, UMR7156, Université de Strasbourg, Strasbourg, France ANTONIO JORDÁN-PLA • Departamento de Bioquímica y Biología Molecular and ERI Biotecmed, Facultad de Biológicas, Universitat de València, València, Spain; Department of Molecular Biosciences, The Wenner-Gren Institute, Stockholm University, Stockholm, Sweden JOHN K. KIM • Life Sciences Institute, University of Michigan, Ann Arbor, MI, USA; Department of Biology, Johns Hopkins University, Baltimore, MD, USA CHRISTIAN KLOSE • Lipotype GmbH, Dresden, Germany SARA A. KNAACK • Wisconsin Institute for Discovery, University of Wisconsin at Madison, Madison, WI, USA PISIWAT KONGSOMBOONVECH • Department of Molecular and Cell Biology, School of Natural Sciences, University of California, Merced, Merced, CA, USA ROMAIN KOSZUL • Institut Pasteur, Department Genomes and Genetics, Groupe Régulation Spatiale des Génomes, Paris, France; CNRS, UMR 3525, Paris, France RACHEL LEGENDRE • Institute for Integrative Biology of the Cell (I2BC), UMR 9198 CEA, CNRS, Université Paris Sud, Orsay, France THIBAUT LÉGER • Mass Spectrometry Laboratory, Institut Jacques Monod, UMR7592, CNRS—Univ Paris Diderot, Sorbonne Paris Cité, Paris Cedex 13, France GAËLLE LELANDAIS • Institut Jacques Monod, CNRS UMR 7592, University of Paris Diderot, Paris, France MATTHEW B. LOHSE • Department of Microbiology and Immunology, University of California, San Francisco, San Francisco, CA, USA MARIA MADRIGAL • Department of Molecular and Cell Biology, School of Natural Sciences, University of California, Merced, Merced, CA, USA CHRISTOPHE MALABAT • Génétique des Interactions Macromoléculaires (UMR3525-CNRS), Institut Pasteur, Paris, France MARTIAL MARBOUTY • Institut Pasteur, Department Genomes and Genetics, Groupe Régulation Spatiale des Génomes, Paris, France; CNRS, UMR 3525, Paris, France JAWAD MERHEJ • Laboratoire de Biologie Computationnelle et Quantitative, Sorbonne Universités, UPMC University of Paris 06, UMR 7238, Paris, France; Laboratoire de Biologie Computationnelle et Quantitative, CNRS, UMR 7238, Paris, France ANA MIGUEL • Departamento de Bioquímica y Biología Molecular and ERI Biotecmed, Facultad de Biológicas, Universitat de València, València, Spain AARON P. MITCHELL • Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, PA, USA PEDRO T. MONTEIRO • INESC-ID, Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Portugal NICOLAS MORIN • INRA, UMR1319 Micalis, Jouy-en-Josas, France; AgroParisTech, UMR Micalis, Jouy-en-Josas, France JULIEN MOZZICONACCI • LPTMC, Université Pierre et Marie Curie, Paris, France OLIVIER NAMY • Institute for Integrative Biology of the Cell (I2BC), UMR 9198 CEA, CNRS, Université Paris Sud, Orsay, France CÉCILE NEUVÉGLISE • INRA, UMR1319 Micalis, Jouy-en-Josas, France; AgroParisTech, UMR Micalis, Jouy-en-Josas, France CLARISSA J. NOBILE • Department of Molecular and Cell Biology, School of Natural Sciences, University of California, Merced, Merced, CA, USA Contributors
- 16. xi MARLENE OEFFINGER • Institut de recherches cliniques de Montréal, Montréal, QC, Canada; Département de biochimie et médicine moléculaire, Faculté de médecine, Université de Montréal, Montréal, QC, Canada; Faculty of Medicine, Division of Experimental Medicine, McGill University, Montréal, QC, Canada ZHENGQING OUYANG • The Jackson Laboratory for Genomic Medicine, Farmington, CT, USA; Department of Biomedical Engineering, University of Connecticut, Storrs, CT, USA CELIA PAYEN • Department of Genome Sciences, University of Washington, Seattle, WA, USA VICENT PELECHANO • European Molecular Biology Laboratory (EMBL), Genome Biology Unit, Heidelberg, Germany LOURDES PEÑA-CASTILLO • Department of Biology, Memorial University of Newfoundland, St. John’s, NL, Canada; Department of Computer Science, Memorial University of Newfoundland, St. John’s, NL, Canada JOSÉ E. PÉREZ-ORTÍN • Departamento de Bioquímica y Biología Molecular and ERI Biotecmed, Facultad de Biológicas, Universitat de València, València, Spain KUN QU • Howard Hughes Medical Institute and Program in Epithelial Biology, Stanford University School of Medicine, Stanford, CA, USA SUSHMITA ROY • Wisconsin Institute for Discovery, University of Wisconsin at Madison, Madison, WI, USA; Department of Biostatistics and Medical Informatics, University of Wisconsin at Madison, Madison, WI, USA ISABEL SÁ-CORREIA • Biological Sciences Research Group, Department of Bioengineering, Instituto Superior Técnico, IBB – Institute for Bioengineering and Biosciences, Universidade de Lisboa, Lisbon, Portugal COSMIN SAVEANU • Génétique des Interactions Macromoléculaires (UMR3525-CNRS), Institut Pasteur, Paris, France JOSEPH SCHACHERER • Department of Genetics, Genomics and Microbiology, CNRS, UMR7156, Université de Strasbourg, Strasbourg, France MARKUS S. SCHRÖDER • School of Biomolecular and Biomedical Science, Conway Institute, University College Dublin, Belfield, Dublin, Ireland EVA SERNA • Servicio de Análisis Multigénico, INCLIVA, Universitat de València, València, Spain NORMA V. SOLIS • Division of Infectious Diseases, Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA, USA KIRILL TARASOV • Department of Biochemistry and Molecular Medicine, Université de Montréal, Montréal, QC, Canada MIGUEL C. TEIXEIRA • Biological Sciences Research Group, Department of Bioengineering, Instituto Superior Técnico, IBB – Institute for Bioengineering and Biosciences, Universidade de Lisboa, Lisbon, Portugal DAWN A. THOMPSON • Broad Institute of MIT and Harvard, Cambridge, MA, USA CHRISTIAN TRAHAN • Institut de recherches cliniques de Montréal, Montréal, QC, Canada; Département de biochimie et médicine moléculaire, Faculté de médecine, Département de biochimie et médicine moléculairel, Montréal, QC, Canada MATHIEU VIDELIER • Mass Spectrometry Laboratory, Institut Jacques Monod, UMR7592, CNRS—Univ Paris Diderot, Sorbonne Paris Cité, Paris Cedex 13, France YUE WAN • Stem Cell and Developmental Biology, Genome Institute of Singapore, Singapore, Singapore CAN WANG • School of Biomolecular and Biomedical Science, Conway Institute, University College Dublin, Belfield, Dublin, Ireland WENJIE XU • Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, PA, USA Contributors
- 18. 1 Frédéric Devaux (ed.), Yeast Functional Genomics: Methods and Protocols, Methods in Molecular Biology, vol. 1361, DOI 10.1007/978-1-4939-3079-1_1, © Springer Science+Business Media New York 2016 Chapter 1 Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species Can Wang, Markus S. Schröder, Stephen Hammel, and Geraldine Butler Abstract The ability to extract, identify and annotate large amounts of biological data is a key feature of the “omics” era, and has led to an explosion in the amount of data available. One pivotal advance is the use of Next- Generation Sequencing (NGS) techniques such as RNA-sequencing (RNA-seq). RNA-seq uses data from millions of small mRNA transcripts or “reads” which are aligned to a reference genome. Comparative transcriptomics analyses using RNA-seq can provide the researcher with a comprehensive view of the cells’ response to a given environment or stimulus. Here, we describe the NGS techniques (based on Illumina technology) that are routinely used for comparative transcriptome analysis of fungal species. We describe the entire process from isolation of RNA to computational identification of differentially expressed genes. We provide instructions to allow the beginner to implement packages in R such as Bioconductor. The methods described are not limited to yeast, and can also be applied to other eukaryotic organisms. Key words Candida, Next-generation sequencing, Illumina, Bioconductor 1 Introduction Transcriptome analysis using RNA-seq quantifies transcription levels across the entire genome [1]. Transcript numbers are mea- sured by sequencing; the number of reads obtained from a specific gene in a test condition compared to a control is used to measure changes in gene expression [2, 3]. Studying the transcriptome in different yeast species has helped elucidate the mechanisms behind key cellular processes and pathways [4–7]. As Next-Generation Sequencing (NGS) technologies drop in price, RNA-seq has become widely used as a method for analyzing gene expression under an array of conditions, and holds many advantages over other similar analytical techniques such as microarrays [2, 8]. Although most of the techniques required can be carried out “in- house,” there are many private companies that now provide NGS services. They will sequence user-provided libraries, but will also
- 19. 2 isolate poly(A) RNA from total RNA preparations and construct libraries, at additional cost. This makes RNA-seq accessible to almost any laboratory. In this chapter, we describe how to carry out RNA-seq analysis from RNA isolation to computational analysis. Labs without access to next-generation sequencing technologies can use commercial companies for the sequencing steps, and move straight to the com- putational analysis section, where the interpretation of results is discussed. We describe a series of tools implemented in the R sta- tistical language [9]. A wide variety of bioinformatics tasks and collections of R packages, such as Bioconductor [10] or CRAN [11] make it possible to utilize R for almost any task associated with analyzing and visualizing sequencing data. We have provided a set of instructions that make it possible for even the beginner to implement tools such as DeSeq2 [12] on a laptop or personal com- puter, to analyze changes in gene expression. 2 Materials 1. NanoDrop spectrophotometer. 2. Agilent 2100 Bioanalyzer. 3. Qubit Fluorometer. 4. Dark Reader-Blue Light Transilluminator. 5. Bead beater. 6. Next-Generation DNA Sequencer (e.g., Illumina platforms Genome Analyzer IIx, HiSeq 2500, or MiSeq), or commercial sequencing services. 7. PC or laptop with Linux or Mac OS X as operating system and Internet access. 1. Yeast RNA Extraction Kit (e.g., Ribopure, Ambion). 2. RNA 6000 Nano Kit (Agilent). 3. High-sensitivity DNA Kit (Agilent). 4. Zinc RNA Fragmentation Kit. 5. Gel Excision Tips (e.g., GeneCatcher). 6. PCR Purification Kits. 7. Qubit dsDNA High Sensitivity Assay Kit (or equivalent). 8. Quick Ligation Kit. 1. 1× Binding Buffer: 20 mM Tris–HCl pH 7.5, 1.0 M LiCl, 2 mM EDTA. 2. 1× Washing Buffer: 10 mM Tris–HCl pH 7.5: 0.15 M LiCl: 1 mM EDTA. 2.1 Specialized Equipment 2.2 Kits 2.3 Buffers and Reagents Can Wang et al.
- 20. 3 3. Tris–NaCl Buffer (50 μl 1 M Tris–HCl pH 7.5, 10 μl 5 M NaCl, and 940 μl Nuclease-free water). 4. 10 mM Tris–HCl, pH 8.5. 5. RNAlater. 6. Dynabeads Oligo (dT)25 (e.g., Dynal from Ambion). 7. dNTP mix (10 mM dATP, dTTP, dCTP, and dGTP). 8. UTP mix (10 mM dATP, dCTP, dGTP, 20 mM dUTP). 9. Reverse transcriptase with buffers (e.g., Superscript III). 10. RNaseOUT. 11. DNA Polymerase I. 12. Klenow DNA Polymerase I. 13. Klenow Fragment (3′→5′ exo-). 14. T4 DNA polymerase. 15. T4 DNA Ligase. 16. T4 polynucleotide kinase. 17. High Fidelity PCR polymerase. 18. Ribonuclease H. 19. Uracil DNA glycosylase. 20. G-50 column (e.g., Illustra Microspin). 3 Methods 1. Inoculate overnight cultures in 5 ml growth medium incubated at a relevant temperature (often 30 °C, shaking at 200 rpm). 2. Sub-culture to an A600nm of 0.2 in 50 ml growth medium and grow to mid-log phase (incubation time depends on growth rates of different species being studied). At this point, an addi- tional treatment can be used, for example treatment with a drug, or a change in temperature or oxygen concentration. 3. Following treatment, retrieve cells from 25 ml culture either by centrifugation for 5 min at 4 °C at 3160×g, or to avoid stress [13] by collection on a filter (0.45 μm nitrocellulose membrane filter) using a vacuum source. 4. Resuspend the cell pellet in 100 μl RNAlater stabilization solu- tion and store at −80 °C until required. The RNAlater solution inactivates any RNases and prevents any changes in expression of the RNA. 1. Treat all lab surfaces and pipettes with 70 % ethanol and an RNase decontamination solution (e.g., RNaseZap) to remove any unwanted RNases. 3.1 Preparation of RNA 3.1.1 Cell Growth 3.1.2 RNA Isolation, Yield, and Quality Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 21. 4 2. Thaw cells on ice. 3. Extract RNA using a commercial kit, following the manufac- turer’s instructions. We use Ribopure Yeast RNA Extraction Kit from Ambion (see Note 1). 4. Determine RNA concentrations below 50 ng/μl by measuring absorbance with a Qubit fluorometer. Use a NanoDrop to identify contaminants [14]. A reading at 260 nm is used to determine concentration (A260 of 1=40 μg/ml). Proteins absorb at 280 nm. The A260/280 ratio therefore provides a mea- surement of the purity of the RNA; the ratio should lie between 1.8 and 2.2. Ethylenediaminetetraacetic acid (EDTA), carbo- hydrates, and phenol all have absorbance near 230 nm. The A260/230 ratio is therefore used as a secondary measure of nucleic acid purity. Expected A260/230 values lie in the range of 2.0–2.2. If the ratio is appreciably lower than this, contaminants are probably present and the sample should not be used. 5. Measure RNA quality with a fluorometric based analytical sys- tems, e.g., Bioanalyzer from Agilent Technologies, following the manufacturer’s instructions. Analysis on a Bioanalyzer gen- erates a graphical visualization in the form of an electrophero- gram of ribosomal peaks (28S and 18S), peak ratio, RNA concentration, a calculated RIN (RNA Integrity Number) value, and a gel-like image of the RNA sample. The RIN value is a measurement of the overall integrity of a given RNA sam- ple that is not affected by sample concentration but by the overall RNA content and background degradation. RIN values >6 are considered to be of acceptable quality. The quantitative range for the RNA 6000 Nano Kit is 5–500 ng/μl. The library protocol described here was developed for sequencing on an Illumina Genome Analyzer IIx. Most analysis is now carried out with the more recent HiSeq and MiSeq systems from the same company. The protocol described here may be adapted for use with the newer platforms (HiSeq/MiSeq/NextSeq) with some minor updates (see Subheading Adapter Synthesis). The steps required for library generation are shown in Fig. 1. This protocol generates strand-specific information by incorporating dUTP during the syn- thesis of the second strand cDNA synthesis [15–17]. This is subse- quently removed by digestion with uracil DNA glycosylase (UDG). There are several variations of the dUTP method, including com- bining with Illumina TruSeq kits [15, 18]. Other methods for gen- erating strand-specific data are described by Levin et al. [16]. 1. Dilute 10 μg total RNA in 50 μl using nuclease-free water. 2. Incubate at 65 °C for 5 min to disrupt RNA secondary struc- tures, and then place on ice. 3.2 Library Generation 3.2.1 Purification of Poly(A) RNA (Fig. 1a) Can Wang et al.
- 22. 5 AAAAAAA TTTTTTT mRNA Oligo(dT)25 beads Poly(A) RNA a d g j k l h i e f b c Zn-mediated Fragmentation mRNA 1st strand cDNA 5’ 3’ Random hexamer primer 5’ 3’ T T T T T 2nd strand cDNA End repair “A” base addition dUTP Adapter ligation Size selection 2nd strand digestion 200 bp 300 bp RNA fragments Library enrichment Library confirmation and purification 200 bp 300 bp Quality control 1. Qubit 2. DNA Chip Concentration Size 10 nM Final library 3’ A 5’ A 5’ 3’ T* *T P P 3’ A 5’ A 5’ 3’ T* *T P5 P7 SR 1.1 SR 1.2 5’ 3’ T T T T T 5’ 3’ T T T T T 5’ 3’ T T T T T 3’ 5’ U UU U U 5’ 3’ T T T T T 3’ 5’ U UU U U 5’ 3’ A T T T T T A 3’ 5’ U UU U U T T T T T U UU U U 3’ A 5’ A 5’ 3’ T* *T T T T T T U UU U U UDG Treatment T T T T T A A A A A Fig. 1 The workflow for constructing strand-specific libraries from total RNA. Each step is described in detail in the text. (a) Poly(A) RNA is selected by binding to oligo(dT)25 Dynabeads. (b) mRNA is fragmented using Zn-mediated fragmentation. (c) First strand cDNA is synthesized using random hexamer primers. (d) Second strand cDNA is synthesized incorporating U instead of T (e) Ends of the cDNA fragments are repaired. (f) “A” bases are added to the 3′ ends of the cDNA fragments (g) Y-shaped iAdapters anneal to the cDNA fragments by overlapping “T” and “A” bases (h) cDNAs ranging from size 200 to 250 bp and 250 to 300 bp are isolated from the gel (shown with arrows). (i) The bottom strand is copied by priming from SR1.2, and the library is amplified using primers SR1.2 and SR1.1 (which adds the P5 sequence, see Fig. 2 for alternatives). (k) Amplification of the library is confirmed by electrophoresis on a 2.5 % agarose gel, and the library (ranging from 200 to 250 bp) is purified from a gel. (l) Library concentration and size are estimated and are diluted to 10 nM for sequencing Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 23. 6 3. Wash 100 μl Dynabeads Oligo (dT)25 with 100 μl 1× Binding Buffer twice using a magnetic rack and resuspend the beads in 50 μl 1× Binding Buffer. 4. Mix 50 μl heated RNA from step 2 and 50 μl washed beads from step 3 and rotate the mixture for 5 min at room tempera- ture. Recover the beads using a magnetic rack and wash twice with 100 μl 1× Washing Buffer. 5. Elute the mRNA in 20 μl Tris–HCl (10 mM, pH 7.5) by heat- ing at 80 °C for exactly 2 min. 6. Wash the beads twice with 1× Washing Buffer. 7. Add 80 μl 1× Binding Buffer to the beads and the 20 μl mRNA from step 5, and repeat the poly(A) selection. 8. Elute the poly(A) RNA in 10 μl 10 mM Tris–HCl (pH 7.5) by heating at 80 °C for exactly 2 min. 9. Recover the RNA from the beads immediately using a mag- netic stand, and transfer 9 μl to thin wall PCR tubes. Store the mRNA at −80 °C. 1. Add 1 μl 10× Fragmentation Buffer (from kit) to 9 μl purified mRNA (poly(A) RNA) in a PCR tube. 2. Incubate the mixture at 70 °C in a thermocycler for 5 min. 3. Add 1 μl Stop Buffer (from kit) and incubate briefly on ice. 4. Add 1 μl 3 M sodium acetate (pH 5.2), 2 μl 5 μg/μl glycogen, and 30 μl 100 % ethanol and precipitate the mRNA at −80 °C for ≥30 min followed by centrifugation at 17,000×g at ≤4 °C for 25 min. 5. Remove the supernatant carefully and wash the pellet with 700 μl 80 % ethanol. 6. Air-dry the pellet and resuspend it in 10.5 μl Nuclease-free water. 1. Add 1 μl of random hexamer primer (3 μg/μl, Invitrogen) to 10.5 μl fragmented mRNA. 2. Incubate the mixture at 65 °C for 5 min and then place on ice. 3. Add 4 μl 5× First Strand Buffer (supplied with reverse tran- scriptase), 2 μl DTT (100 mM), 1 μl 10 mM dNTP mix, and 0.5 μl RNase OUT (40 units/μl), incubate at 25 °C for 2 min and then add 1 μl of Reverse Transcriptase (Superscript III is recommended) to each sample. 4. Incubate the mixture at 25 °C for 10 min, 42 °C for 50 min and then 70 °C for 15 min. 5. Store the first strand cDNA on ice. 3.2.2 Zinc-Mediated Fragmentation of mRNA (Fig. 1b) 3.2.3 First Strand cDNA Synthesis (Fig. 1c) Can Wang et al.
- 24. 7 6. Remove dNTPs and hexamers by centrifugation through a G-50 spin column. Centrifuge the G-50 column at 2000×g for 1 min. Add the first strand cDNA sample carefully to the top and center of the resin and collect by centrifuging for 2 min at 2000×g. 7. Immediately carry out second strand cDNA synthesis. 1. Incubate all reagents on ice for 5 min prior to use. 2. Add 1.3 μl 5× First Strand Buffer, 20 μl 5× Second Strand Buffer (supplied with reverse transcriptase), 3 μl dUTP mix (10 mM dATP, dCTP, dGTP, 20 mM dUTP), 1 μl DTT (100 mM), 5 μl E. coli DNA Polymerase I (10 units/μl), and 1 μl Ribonuclease H (2 units/μl) to the first strand samples. 3. Add Nuclease-free water to bring the volume to 100 μl. 4. Incubate at 16 °C for 2.5 h, and purify the cDNA in 30 μl 10 mM Tris–HCl, pH 8.5 or equivalent solution supplied with PCR purification kit as per manufacturer’s guidelines, and store at −80 °C. Treating the cDNA fragments with a combination of T4 DNA polymerase and E. coli DNA polymerase I Klenow fragments removes 3′ overhangs via the 3′–5′ exonuclease activity, while the polymerase fills in any 5′ overhangs. Both these steps are necessary to facilitate ligation of sequencing adaptors. A single adenosine base is added to the 3′-end of the cDNA fragments to facilitate ligation to the sequencing adapter. 1. Add 45 μl Nuclease-free water, 10 μl T4 DNA Ligase buffer with 10 mM ATP, 4 μl dNTP mix (10 mM), 5 μl T4 DNA polymerase (3 units/μl), 1 μl Klenow DNA Polymerase (5 units/μl), and 5 μl T4 Polynucleotide Kinase (10 units/μl) to 30 μl cDNA. Incubate at 20 °C for 30 min, and purify by elution with a PCR purification kit as per manufacturer’s guidelines. This is a safe stopping point, and samples may be stored at −80 °C. 2. To add an A base to the 3′ end, add 5 μl Klenow buffer, 10 μl dATP (1 mM), and 3 μl Klenow Exo Fragment (5 units/μl , 3′→5′ exonuclease) to end repaired cDNA, in a total volume of 50 μl. 3. Incubate the reaction at 37 °C for 30 min and purify by elution using a PCR purification kit as per manufacturer’s guidelines. Samples with an A overhang should not be stored for long periods as they are unstable. Several libraries can be pooled together and sequenced on the same run. This is achieved by ligating specific adapters containing different barcode (or index) sequences, to the DNA fragments. 3.2.4 Second Strand cDNA Synthesis with dUTP (Fig. 1d) 3.2.5 End Repair (Fig. 1e) and Addition of a Single “A” Base (Fig. 1f) 3.2.6 Adapter Synthesis Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 25. 8 The barcodes are used to separate library specific data after sequencing [19]. We originally used home-made single read (SR) Y-shaped adapters with short six nucleotide barcodes, designed by Dr. Amanada Lohan UCD, and based on the 2008 Illumina cus- tomer letter and Craig et al. [19, 20] (Fig. 2a). These adapters are made from two single stranded oligonucle- otides (Oligo-1 and Oligo-2) with both complementary regions and noncomplementary regions that when annealed create a Y-shaped adapter bound together at the hinge (complementary) region (Fig. 2a). The top oligonucleotide (Oligo-1) contains a T overhang with a phosphorothioate linkage required for stabiliza- tion and resistance to nuclease digestion, that is designed to ligate to the A overhang added to the insert DNA during library genera- tion. Oligo-2 is phosphorylated at the 5′ end during synthesis, and is complementary to the P7 region, which anneals to sequences on the flowcell. An equivalent P5 region is added at the opposite end during library amplification (step 10, Fig 1j). The 6 nucleo- tides barcode (index) is added to both Oligo-1 and Oligo-2. Table 1 shows six barcode indexes, allowing six libraries to be combined in a single lane (multiplexing). The choice of barcode depends on the number of libraries combined (Table 2). It is now possible to design and synthesize long adapter sequences, using updated recommendations from Illumina [21] that remove the necessity to add the P5 region during library amplification (Fig. 2b, [21]) (see Note 2). 1. Synthesize the oligonucleotides commercially, and using HPLC purification. To construct the Y-shaped adapters, resus- pend lyophilized oligonucleotides in 10 mM Tris–HCl at 100 pmol/μl and anneal them together to form a forked adapter (iAdapter), making sure that each oligo contains the same barcode/index sequence. 2. Add 20 μl of the relevant indexed Oligo-1, 20 μl indexed Oligo-2 and 10 μl Tris–NaCl Buffer in 0.2 ml PCR tube. Fig. 2 (continued) of the barcode index (using index sequence of iSR-6 (Table 1) as an example). The inserted cDNA is shown in italics.The P7 sequence is highlighted in dark grey. Before library amplification the first (top) strand (contain U residues) is degraded, and the bottom strand is copied using SR1.2 as a primer. Subsequent amplification with primers SR1.1 and SR1.2 adds the P5 sequence (light grey). (b) Library generation using updated Illumina recommendations [21].Two oligonucleotides (the universal adapter and indexed adapter) are synthesized for each library, and a Y-shaped adapter is generated as in 2A. The cDNA is ligated at the arrow. The universal adapter sequence contains the P5 sequence and the indexed adapter contains the P7 sequence. The P7 sequence also contains an index/barcode (In). The libraries are amplified with primers 1 and 2. The number of multiplexed samples can be increased by also including indexes in the universal primer (dual indexing, not shown). For single reads a sequencing primer for the P7 end is used, for paired-end reads, primers from both P7 and P5 ends are used. Advice on designing and synthesizing longer adapters is available from refs. [15, 45]. The asterisk indicates a phosphorothioate linkage, and the P indicates a phosphorylated nucleotide Can Wang et al.
- 26. 9 a 5’ACACTCTTTCCCTACACGAC GCTCTTCCGATCAGCTAT*T -NUN...NUN.A ATAGCTGATCGGAAGAGC CGAGAAGGCTAGTCGATA A..NTN..NTN..T*TATCGACTAGCCTTCTCG 3’GTTCGTCTTCTGCCGTATGCT TCGTATGCCGTCTTCTGCTTG 3’ CAGCACATCCCTTTCTCACA 5’ Oligo-1 cDNA fragment P P Oligo-2 5’AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGAC 3’GTTCGTCTTCTGCCGTATGCTCTA(In)CACTGACCTCAAGTCTGCACA GCTCTTCCGATC*T CGAGAAGGCTAG cDNA fragment P P GATCGGAAGAGC T*CTAGCCTTCTCG ACACGTCTGAACTCCAGTCAC(In)ATCTCGTATGCCGTCTTCTGCTTG 3’ CAGCACATCCCTTTCTCACATCTAGAGCCACCAGCGGCATAGTAA 5’ Universal Adapter Indexed Adapter 5’CAAGCAGAAGACGGCATACGAGCTCTTCCGATCAGCTATTNTN...NT.AATAGCTGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT 3’ 3’CTAGCCTTCTCGCAGCACATCCCTTTCTCACA SR 1.1 (P7) cDNA fragment 3’GTTCGTCTTCTGCCGTATGCTCGAGAAGGCTAGTCGATAA..NTN..NTTTATCGACTAGCCTTCTCGCAGCACATCCCTTTCTCACA 5’ 5’CAAGCAGAAGACGGCATACGAGCTCTTCCGATC SR 1.2 T C T A G A G C C A C C A G C G G C A T A G T A A 5 ’ (P5) (P7) b 5’CAAGCAGAAGACGGCATACGAGAT (P7) (P5) Primer 2 5’CAAGCAGAAGACGGCATACGAGAT (Index)GTGA...:cDNA:...TCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT 3’ 3’GTTCGTCTTCTGCCGTATGCTCTA (Index)CACT...:cDNA:...AGCACATCCCTTTCTCACATCTAGAGCCACCAGCGGCATAGTAA 5’ (P7) (P5) AGCACATCCCTTTCTCACATCTAGAGCCACCAGCGGCATAGTAA 5’ Primer 1 Primer 2 SR 1.2 Fig. 2 Adapters used for library generation for Illumina sequencing. (a) iAdapters generated using the original legacy SR (single read) adapters described in the Illumina customers letter prior to 2009.Two oligonucleotides, Oligo-1 and Oligo-2 are synthesized for each adapter.The underlined six nucleotide sequences show the location Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 27. 10 3. Incubate at 97 °C for 2 min, followed by a stepdown of −1 °C/min for 72 cycles, and finally at 25 °C for 5 min. 4. Store the 40 μM iAdapter master stock at −20 °C. For most RNA-seq libraries a 15 μM working stock is used. However when <1 μg starting RNA is available further dilution may be required. 1. Add 25 μl 2× Quick DNA ligase buffer, 1 μl iAdapter mix (15 μM), and 2 μl Quick T4 DNA ligase (NEB) to 22 μl end- repaired cDNA with an A overhang (from the end repair step). 2. Incubate at 20 °C for 15 min and purify by elution in 10 μl 10 mM Tris–HCl, pH 8.5 using a PCR purification kit. 3. Store at −80 °C. 1. Separate samples on a 2.5 % agarose gel by prepared using ultrapure TAE buffer and high-resolution agarose contain- ing Ethidium Bromide at a final concentration of 1 μg/μl (see Note 3). Use a suitable DNA ladder (see Note 4). 3.2.7 Adapter Ligation (Fig. 1g) 3.2.8 Gel Purification (Fig. 1h) Table 2 Pooling strategies Number of libraries Best combinations 2 in the lane (iSR 6, iSR 20) or (iSR 10, iSR13) 3 in the lane (iSR 10, iSR 11, iSR13) or (iSR 6, iSR16, iSR 20) 5 in the lane (iSR 6, iSR 10, iSR 13, iSR 16, iSR 20) 6 in the lane (iSR 6, iSR 10, iSR 11, iSR 13, iSR 16, iSR 20) Table 1 Barcode sequences Adapter IDa Index/barcodeb iSR-6 AGCTAT iSR-10 CGATCT iSR-20 GATCGT iSR-11 GCTAGT iSR-13 TAGCTT iSR-16 TCGATT a SR indexed adapters are designed by adding a barcode of six nucleotides, which have to maintain color balance for each base. A/C bases are identified by the red laser and G/T bases by the green laser on a Genome Analyzer IIx. We show six adapters designed according to Illumina indexing guidelines [21]. Many more are possible (for example, see ref. [45]) Can Wang et al.
- 28. 11 2. Add 10 μl adapter-ligated cDNA to 6 μl gel loading dye (such as Orange G from Promega) and add to the same volume of DNA ladder. 3. Load the wells of the gel with this solution very slowly to pre- vent overflow and spilling. 4. Electrophorese at 80 V for 3 h until sufficient separation of the 100 and 200 bp bands of the DNA ladder has occurred. 5. Visualize the DNA on a Dark Reader Transilluminator, which operates at a wavelength that does not damage the sample, unlike normal UV transilluminators. 6. Excise regions corresponding to 200–250 bp and to 250– 300 bp (for backup) using Gel Excision Tips (for example from GeneCatcher). 7. Elute the adaptor-ligated cDNA in 30 μl of 10 mM Tris–HCl, pH 8.5 by using a Gel Extraction kit. Store the samples at −80 °C. 1. Aliquot 26 μl gel purified adapter-ligated material (200– 250 bp) to sterile 200 μl PCR tubes. 2. Add 3 μl 10× uracil DNA glycosylase (UDG) buffer and 1 μl uracil DNA glycosylase (1 unit/μl). 3. Incubate in a thermal cycler at 37 °C for 20 min. 4. Terminate the reaction by heating at 94 °C for 10 min and 4 °C for 5 min. Store the digested DNA samples at −80 °C. The adaptor-ligated cDNA samples are amplified by PCR to ensure there is a sufficient quantity for sequencing. 1. Add 10 μl 5× buffer (provided with enzyme), 0.8 μl PCR primer SR 1.1 (Fig. 2a), 0.8 μl PCR primer SR 1.2 (Fig. 2a) or other suitable primers (Fig. 2b), 0.8 μl dNTP mix (25 mM), and 0.8 μl High Fidelity polymerase (e.g., cloned Phusion polymerase from NEB) to 20 μl digested DNA and bring the total volume to 50 μl with Nuclease-free water. If a different polymerase is used, ensure that it is not inhibited by dUTP. 2. Amplify at 98 °C for 30 s, followed by 12–14 cycles of 98 °C for 10 s, 65 °C for 30 s, 72 °C for 30 s, and then 72 °C for 5 min. In the method shown in Fig. 1j and Fig. 2a the P5 sequence, which is required to hybridize to the sequence on the flow cell, is added to the end of the adapter by amplifica- tion with oligonucleotide SR 1.1. The second oligonucleotide SR 1.2 contains the P7 sequence. Adapters shown in Fig. 2b can be amplified using the oligonucleotide primers shown. 3. Visualize library quality using 1.5 μl of the amplified DNA reaction on a 1 % agarose gel (Fig. 1k). The adapter/dimers should be 100–120 nucleotides long. 3.2.9 Second Strand Digestion (Fig. 1i) 3.2.10 Amplification of Adapter Ligated DNA Templates (Fig. 1j) Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 29. 12 4. Purify the remaining 48.5 μl amplified cDNA by elution in 10 μl 10 mM Tris–HCl, pH 8.5 with a PCR purification kit if a product is visible. 5. Separate the DNA library samples on a high-resolution grade 2.5 % agarose gel. Visualize using a Dark Reader Transilluminator and excise fragments in the range of 200–250 bp, as described in the gel purification step (see Note 3). Store the samples at −80 °C. 1. Quantify the amplified library samples using a Qubit Fluorometer and the Qubit High-sensitivity dsDNA assay, as per manufacturer’s guidelines. 10 nM library dilution is typical for starting point dilution for cluster generation. 2. Check the quality of cDNA library using a High-sensitivity DNA chip assay on a Bioanalyzer, as per manufacturer’s instructions. 1. Normalize cDNA libraries to 10 nM based on Qubit and DNA chip values. Ensure that the libraries contain a single peak of approximately 200 nucleotides, with little or no evidence of adapter dimers. 2. Dilute the library with 10 mM Tris–HCl, pH 8.5 with 0.1 % Tween 20 recommended for stability of library. Add 10 μl of each adapter-ligated library per lane. It is recommended that only certain barcodes (indexes) are combined together (Table 2). A minimum of 10 μl (one library) is required for clustering process. We carry out cluster generation and sequencing using an in-house Genome Analyzer IIx platform (see Note 5). The multi-indexed library mix is loaded on the Illumina 8 channel flowcell. For the first step cluster generation, hundreds of millions of templates are hybridized to a lawn of oligo nucleotides immobilized on the flow cell surface. Immobilized DNA template copies are amplified by isothermal bridge amplification. The process is repeated on each template by cycles of isothermal denaturation and amplification to create millions of individual copies. Each cluster of dsDNA bridges is denatured and reverse strand is removed by specific base cleavage, leaving the forward DNA strand. After strand blocking on the flow- cell surface, the sequencing primer is hybridized to the complemen- tary sequence on the adapter on unbound ends of the templates in the clusters and each cycle of sequencing identifies a single base. A substantial part of RNA-seq experiments consists of computa- tional processing and analysis of the data. These analyses range from filtering the raw reads obtained from the sequencing machine, to differential gene expression analysis and biological interpreta- tion of the results. 3.2.11 Quality Control and Purification of Final Library (Fig. 1l) 3.2.12 Pooling Strategy 3.2.13 Cluster Generation and Sequencing 3.3 Prerequisites to Computational Analysis Can Wang et al.
- 30. 13 In the following sections we describe how to download, process, and analyze RNA-seq data using Mac OS X or a Linux distribution (such as Ubuntu) as the operating system. A server or computer cluster (e.g., Amazon EC2) can also be used. To illustrate the use of the software we use a subset of recently published data from an experiment investigating the differences between the transcriptome of Candida parapsilosis grown as bio- films and under planktonic growth conditions (Table 3) [22]. This is strand-specific transcriptional profiling data obtained from a commercial company (BGI, Hong Kong) using an Illumina HiSeq 2000 with paired end reads of 90 bases. We describe in some detail how to visualize the results using a combination of R [9] and Bioconductor [10]. 1. Download the dataset from the Gene Expression Omnibus (GEO [23]) under GEO accession number GSE57451. It includes wild type C. parapsilosis cultures grown under plank- tonic and biofilm conditions. The individual reads are stored in the Sequence Read Archive (SRA [24]) under accession num- ber SRP041812. Download the data using your favorite tool, or use the Unix command “wget” to import the files to your server or hard drive. The total size is 10 GB. 2. Alternatively, all required data and output files generated by working through the exercises in the computational analysis section can be downloaded from http:/ /www.cgob.ie/supp_ data. The directory structure follows the Unix setup described under “Common Unix Setup”, which makes it easy to verify that your locally generated results are correct. 1. Use a structured directory system for all sequencing related software and data to help to keep track of files and installed software. The setup described here is applicable for both Mac OS X and Linux operating systems. In Mac OS, use the 3.3.1 Download Data Set 3.3.2 Common Unix Setup Table 3 Sequencing Read Archive accession numbers for samples used as an example SRA accession number Description SRR1278968 C. parapsilosis planktonic replicate 1 SRR1278969 C. parapsilosis planktonic replicate 2 SRR1278970 C. parapsilosis planktonic replicate 3 SRR1278971 C. parapsilosis biofilm replicate 1 SRR1278972 C. parapsilosis biofilm replicate 2 SRR1278973 C. parapsilosis biofilm replicate 3 Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 31. 14 “Terminal” application to execute commands and manipulate files on the hard drive. For the purpose of the RNA-seq work- flows below, create a general folder called “ngs” and subfolders for “data” and “applications”. If you are using a server and do not have root access, the folder “~/local” and “~/local/bin” should be created as well. The “~/” represents the home direc- tory and “~/local/bin” is the general location where executa- bles and file links (Unix command “ln –s”) to applications are stored. Use the following commands create the directories: mkdir ~/ngs mkdir ~/ngs/data mkdir ~/ngs/applications mkdir ~/local mkdir ~/local/bin 2. Extend the PATH variable to include the “~/local/bin” folder. This enables the execution of the installed software anywhere on the system by typing the name of the software, rather than the full path to the directory where the software is installed. Use the following commands to extend and view the PATH variable: export PATH=$HOME/local/bin:$PATH echo $PATH This change to the PATH variable is temporary and will be lost after logging out of the current session. To permanently extend the PATH variable add the command “export PATH=$HOME/local/bin:$PATH” to the end of either the “.profile” or “.bashrc” file in the home directory using for example emacs or vim. 3. Most software can be executed directly after unpacking down- loaded archives. More advanced users, or users working with operating systems other than Linux or Mac OS X can build software from source (see Note 6). Use the following commands: – Downloading files to a server/local hard drive from the command line: wget http:/ /some.web.address/file.tar.gz. – Unpacking an archive: tar xvfz file.tar.gz. – Creating a folder: mkdir name-of-folder. – Configuration of the installation script: ./configure --prefix=$HOME/local/bin. – Building the software: make; make test; make install. – Linking the new software to a folder that is included in the $PATH variable: ln -s $PWD/new-software-executable ~/local/bin/ Can Wang et al.
- 32. 15 If a program depends on external tools that need to be installed, the README or INSTALL files from the downloaded archive provide further details. All software required for the analysis workflow under “Data pro- cessing” are listed in Table 4. After downloading the individual files, execute the commands below inside the Terminal application in Mac OS X or Linux. 1. SRA Toolkit is available as compiled binaries. Download the archive into “~/ngs/applications”, unpack, and link the exe- cutable to “~/local/bin”. For the Ubuntu SRA Toolkit ver- sion 2.3.5, use the commands: tar xvfz sratoolkit.2.3.5-2-ubuntu64.tar.gz cd sratoolkit.2.3.5-2-ubuntu64/bin ln -s $PWD/fastq-dump ~/local/bin/ 2. SAMtools (version 1.0 and above) is available as compiled binaries for Linux and Mac OS X that include SAMtools, BCFTools and HTSlib. The current version of SAMtools (v1.1) is not yet compatible with TopHat and we therefore recommend using SAMtools v0.1.19. This issue might be fixed with a TopHat version above 2.0.12. Download SAMtools to the applications folder and execute the following commands (for SAMtools version 0.1.19): tar xvfz samtools-0.1.19.tar.bz2 cd samtools-0.1.19/ make ln -s $PWD/samtools ~/local/bin 3. FastQC is based on Java and platform independent. Java is installed by default on current Linux and Mac OSX operating systems. To test which version of Java is installed execute: java -version Install the Mac OS X version of FastQC by copying the FastQC bundle into the applications folder. Download the Linux ver- sion into the applications folder and execute (for FastQC ver- sion 0.11.2): unzip fastqc_v0.11.2.zip cd FastQC chmod 755 fastqc ln -s $PWD/fastqc ~/local/bin The “chmod” command changes the “fastqc” file permission to make it executable. 3.3.3 Installing Software Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 33. Table 4 Overview of file formats, software, and resources used for processing and analyzing RNA-seq data Format Description SRA Used by the Sequence Read Archive to store and provide sequencing data FASTQ For storing sequencing data and corresponding quality scores GTF/GFF General Feature Format/Gene Transfer Format. Standardized formats for storing gene information SAM Sequence Alignment Map. Tab-delimited file format for storing alignment information from sequencing reads BAM Binary version of SAM format with a significantly smaller file size BED Format to store specific meta-data for regions of the genome. Used by the UCSC Genome Browser BEDGRAPH Based on the BED format. Stores scored data for specified genomic regions BIGWIG Very large collections of the BEDGRAPH format can be transformed into the binary BIGWIG format Software/resources Reference/website Description SRA/SRA Toolkit [24] ncbi.nlm.nih.gov/Traces/sra Storage for raw sequencing reads Collection of scripts for handling SRA files SAMtools [37] www.htslib.org/ Collection of scripts to handle SAM files Skewer [30] sourceforge.net/projects/skewer Tool for quality trimming and filtering of sequencing reads FastQC [29] bioinformatics.babraham.ac.uk/ projects/fastqc Generates quality reports for sequencing files Bowtie2 [36] bowtie-bio.sourceforge.net/ bowtie2 Tool for aligning sequencing reads to a reference genome TopHat [31] ccb.jhu.edu/software/tophat Splice-aware aligner for sequencing reads to a reference genome HTSeq [38] www-huber.embl.de/users/ anders/HTSeq Python based tools to analyze sequencing data. The script htseq-count calculates read counts per gene R [9] www.r-project.org Statistical language used for a variety of computational biology tasks Bioconductor [10] www.bioconductor.org Large collection of R packages for biological data CRAN [11] cran.r-project.org Large collection of R packages CGD [40] www.candidagenome.org Extensive resource for Candida species IGV [25] www.broadinstitute.org/igv Integrative Genomics Viewer for displaying sequencing data Cytoscape [64] www.cytoscape.org Open source platform for visualizing and analyzing network data Python [65] www.python.org Programming language commonly used for Bioinformatics tasks
- 34. 17 4. Skewer is available as compiled binaries for Mac OS X and Linux. Download the respective binary and execute the fol- lowing commands: mkdir skewer mv skewer-0.1.118-linux-x86_64 skewer/ cd skewer chmod +x skewer-0.1.118-linux-x86_64 ln -s $PWD/skewer-0.1.118-linux-x86_64 ~/local/bin/skewer 5. For both MacOS and Linux, download and unpack the Bowtie2 archive and link “bowtie2” and the genome indexer “bowtie2- build” to “~/local/bin” with the commands (for version 2.2.3): unzip bowtie2-2.2.3-linux-x86_64.zip cd bowtie2-2.2.3 ln -s $PWD/bowtie2 ~/local/bin ln -s $PWD/bowtie2-build ~/local/bin 6. Installation of TopHat is very similar to bowtie2 with the fol- lowing commands (for version 2.0.12): tar xvfz tophat-2.0.12.Linux_x86_64.tar.gz cd tophat-2.0.12.Linux_x86_64 ln -s $PWD/tophat ~/local/bin 7. Install Python (version number above 2.5 and below 3.0) before installing HTSeq. Most servers and computer clusters will have one or several versions of Python already installed. Check the version of Python using: python --version If Python is not installed, or installed with a wrong version number, use the Unix tool “apt-get” to install Python. The user needs to have “sudo” rights for the following command to install Python 2.7, and the two additional packages that HTSeq requires, numpy and matplotlib: sudo apt-get install build-essential python2.7-dev python- numpy python-matplotlib If no “sudo” rights are available and Python or the numpy or matplotlib packages are not installed, contact the systems administrator to install them. To verify that numpy and mat- plotlib are installed, execute the following commands (the “python” command will enter the Python command line): python import numpy import matplotlib If no error messages are displayed the packages are installed and ready to use (see Note 7). Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 35. 18 8. To install HTSeq download and unpack the HTSeq archive and install it using Python with the following commands (for version 0.6.1): tar xvfz HTSeq-0.6.1.tar.gz cd HTSeq-0.6.1 python setup.py install –user ln –s $PWD/build/scripts-2.6/htseq-count ~/local/bin 9. For Mac OS X, an installation package for R is provided. Download the newest version, double-click the downloaded file and follow the instructions of the Mac OS X installer. On Linux systems execute the command: sudo apt-get install r-base r-base-dev If no “sudo” rights are available, download the source package of R, unpack the archive and build R with the following commands: tar xvfz r-base_3.1.1.orig.tar.gz cd R-3.1.1 ./configure –-prefix=$HOME/local/bin make ln –s $PWD/bin/R ~/local/bin 10. The Integrative Genomics Viewer [25] can be downloaded and used locally or launched directly from a web browser with varying amounts of allocated memory. 11. The Bioconductor package DESeq2 is required for the differ- ential expression analysis described under “Generating HTML reports”. The Bioconductor package ReportingTools is used to create HTML reports from DESeq2 results. Start R and execute the following commands: source("http://bioconductor.org/biocLite.R") biocLite("DESeq2") biocLite("ReportingTools") Several different file formats are required. The user should become familiar with the various types listed in Table 4 and described below. SRA: File format used by the Sequence Read Archive [24] to store and provide sequencing data. SRA format files can be converted into several commonly used formats using SRA Toolkit. SRA files in the sequence read archive format (file ending “.sra”) can be transformed into the FASTQ format using “fastq-dump” from SRA-tools (see Note 8). FASTQ: A text based format for storing sequencing data and qual- ity scores. Each entry (read) in the FASTQ format consists of four lines. These represent (1) sequence identifier and description, 3.3.4 File Formats Can Wang et al.
- 36. 19 (2) the sequence, (3) an optional line that starts with a “+” and most commonly includes the sequence identifier and description again and (4) the Phred scale that is used to measure the base qual- ity. Phred scores indicate the probability of incorrect base calls and the Phred scale is based on ASCII characters. For current Illumina sequencing data the ASCII encoded scores have an offset of 64 and raw base qualities normally range from character @ (quality 0) to i (quality >40). A shortened example of one FASTQ file entry is shown below: 1. @SRR1278968.1.1 FCC1WYWACXX:1:1101:1238:2126 length=90 2. TGGGNCTGTACGTGGTTCTTCAATTGCTTGTTTGTT CAATGGTAAATTCG[…] 3. +SRR1278968.1.1 FCC1WYWACXX:1:1101:1238:2126 length=90 4. ___cBQaccgg^ee[ddeegghfff`gghbe_cegffaa_c^_aeedc`[…] GTF/GFF: The General Feature Format (GFF) and Gene Transfer Format (GTF) are two very similar formats used to store feature (gene) information. These include the genomic locations of exons, Coding Sequences (CDS), transcripts, 3′ and 5′ UnTranslated Regions (UTRs), tRNA, etc. The GFF file for an organism is used to assign features to sequencing reads that are mapped to the genome. An example of a line from a C. parapsilosis GFF file is shown below: Cp_c1 . exon 94585 95295 . - . gene_id "CPAR2_100565_ exon"; transcript_id "CPAR2_100565_mRNA" It stores the chromosome name (Cp_c1), feature type (exon), start and stop positions, strand (-), as well as additional attri- butes that are used for feature annotation, for example the transcript name (CPAR2_100565_mRNA). Single dots “.” in this example indicate missing/empty information (see Note 9). SAM: The Sequence Alignment/Map (SAM) format is a tab- delimited file to store alignment information for sequencing reads. There are eleven mandatory columns for each entry, which include information such as the sequence identifier, a bitwise FLAG that provides a summary of the read alignment, mapping position and quality score of the mapping. Additional columns can contain more specific information, such as the number of times the sequence mapped to the genome (NH), comments (CO), or mate pair information if the sequence data is generated from paired-end reads (MC, MQ). An example of a line from a SAM file is shown below: HWI-D00382:125:C48G6ACXX:8:1101:1134:59125 137 Cp_c8 2005578 50 101M * 0 0 AGCTGGTATCTTGTTG ACCCCAACTTTTGTCAAGTTGATTGCTTGGTACGATA Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 37. 20 ACGAATACGGTTACTCCACCAGAGTTGTTGATTT GTTGGAAAAATTTG CCCFFFDEHHHGHIGHHGGIID: CGHEHHGHFGEH>HEHIGIIIHEGHIG=FHGIIG= ACGHEHAAH;C==BBDE(.(6>A?B@;A@CACAA3(:<?B@C4 AS:i:0 XN:i:0 XM:i:0 XO:i:0 XG:i:0 NM:i:0 MD:Z:101 YT: Z:UU XS:A:+ NH:i:1 From the beginning it lists the sequence ID, FLAG, chromo- some name, leftmost mapping position, mapping quality, CIGAR string for the alignment (101 matching bases), sequence ID of mate or read pair (“*” means information unavailable), position of mate, observed template length, raw sequence, and Phred-scaled base quality. Information about the optional fields, such as AS or XN, are available from SAMtools’ GitHub repository [26]. BAM: Smaller binary version of SAM format, can be viewed using the “samtools view” command. BAM files are commonly used to display alignment data in genome browsers because of their smaller size compared to the non-binary SAM format. BED: Mostly used for displaying genomic data in a genome browser. Three fields specify the chromosome, start and end position. Nine additional fields can be used to provide more specific values for the genomic location, i.e., name, score, strand, thickStart, thickEnd, itemRgb, blockCount, block- Sizes, and blockStarts. An example of a bed file generate from TopHat showing deletions found in RNA-seq data compared to the reference genome is below: track name=deletions description="TopHat deletions" Cp_c1 1474 1475 - 1 Cp_c1 1509 1511 - 2 Cp_c1 1771 1772 - 1 BEDGRAPH: More specific format for displaying continuous- valued data in a genome browser. Based on the BED and WIG formats, the BEDGRAPH format can be used to display con- tinuous-numeric values for genomic regions, for example tran- scriptome data. An example of a BEDGRAPH file is below. The columns represent chromosome name, start and stop position, and the numeric value, e.g., a user-defined score or coverage information. Cp_c1 665 756 -2 Cp_c1 1039 1042 -1 Cp_c1 1042 1067 -2 5. BIGWIG: For very large collections of data the BEDGRAPH files can be converted into the binary BIGWIG format to save disk space. Can Wang et al.
- 38. 21 After successfully installing the required software listed in Table 3 on Linux or Mac OS X, all further commands can be executed in the Terminal. Throughout the workflow below, we will provide example commands for sample SRR1278968 (Table 3). 1. To successfully execute downstream analyses these commands need to be executed separately for all six samples downloaded from SRA (Table 3). This data is already stored in separate files for each sample. If multiple samples are sequenced in the same lane on a sequencing machine, e.g., an Illumina HiSeq 2500, the raw sequencing reads must be separated using the bar- code/indexing information, which is usually the first six bases of each read. The fastx_barcode_splitter tool from the FastX- Toolkit can be used to achieve this [27, 28]. The data in Table 3 was obtained from strand-specific 90 base paired-end sequencing. For each sample there are two different files, one containing the first read of the pair and one containing the second read. The standard file naming convention for paired-end reads ends is “_1” and “_2. Generate the files by providing fastq-dump with the option “--split-3” as shown here: fastq-dump --split-3 SRR1278968.sra 2. To confirm that the conversion from SRA to FASTQ was suc- cessful, use the Unix commands “head” or “less” to briefly examine the generated files. Each sample should have two additional files with the endings “_1.fastq” and “_2.fastq”. The first read in the FASTQ file looks like this: @SRR1278968.1 FCC1WYWACXX:1:1101:1238:2126 length=90 TGGGNCTGTACGTGGTTCTTCAATTGCTTGTTTGT TCAATGGTAAATTCGAGTCATCATGATGTGTTGGAGT TTGATTGGTGATTGTTTG +SRR1278968.1 FCC1WYWACXX:1:1101:1238:2126 length=90 ___cBQaccgg^ee[ddeegghf f f`gghbe_cegf faa_c^_ aeedc`Xe^aeebfa]begbebbZc_bcgR`^`V^R^__]]bB 3. Once all the files are generated, check the overall quality of the data using FastQC, a java based tool that creates extensive summary reports. Use the following commands: mkdir qc fastqc SRR1278968_1.fastq -o qc 1>qc/SRR1278968_1.log 2>qc/SRR1278968_1.err fastqc SRR1278968_2.fastq -o qc 1>qc/SRR1278968_2.log 2>qc/SRR1278968_2.err 3.4 Data Processing 3.4.1 Quality Control and Trimming of Raw Data Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 39. 22 The “mkdir” command creates the “qc” folder where the results will be stored. For each file, FastQC generates a fastqc_ report.html file, which can be displayed in any available browser. The quality report provides a basic summary of the sequencing reads in each file, overall per base and per sequence quality scores of a random subsample of all reads and several statistics to assess the quality of the sequencing run and the sequenced material. FastQC also reports basic sequence analy- sis results, such as over-represented sequences and relative kmer enrichments. To assess of the quality of the sequencing data, it is helpful to look at the per base sequence quality boxplot. Figure 3 shows plots from two different fastq files, one from the sample data. The data in Fig. 3a is of a high quality. The boxplot shows that the quality scores from the base calling rarely fall below 30 for all 90 bases in the reads. Applying quality filtering on this sample will result in discarding a very small number of reads. Figure 3b shows an example of a sample with reads of mixed quality. The black boxes for the first 80 bases are close to or above a base quality of 30, which indicates that the majority of reads has a high quality. However, a large number of reads in the lower 25th percentile fall below an acceptable base quality threshold (e.g., [15]) and have to be trimmed or removed. The quality of reads from a sequencing experiment can vary significantly, the reasons vary from poor quality (degraded) or contaminated starting material to mistakes during the sequencing run itself (e.g., temporary shortage of solutions in the sequencing machine, or bubbles in the flowcell) [29]. 4. After inspecting the FastQC report, trim the raw reads using Skewer [30] (see Note 10). Trimming sequencing data is an important step to ensure that only high quality data is analyzed and the results are not influence by poor quality reads. Skewer was developed primarily to improve adapter trimming of next- generation sequencing data, but it is also one of the fastest tools to remove poor quality bases from paired-end RNA-seq reads. It can utilize multiple processors to further speed up the quality trimming [30]. To run Skewer a few options must be specified. These include “-m pe” for paired-end trimming. Additionally recom- mended thresholds for trimming are a minimum read length of Fig. 3 (continued) and below each represent 25 %. Light, medium, and dark grey background colors indicate poor, medium, and good per base quality, respectively. Quality scores are encoded in Illumina 1.5 format (a) and >1.3 format (b). Expected quality for raw sequencing data in both formats ranges from 0 to 40, with the exception that in Illumina format version 1.5 and above the quality score 2 represents the Read Segment Quality Control Indicator and 0 and 1 are unused Can Wang et al.
- 40. 23 40 Quality socres (Illumina 1.5) a b Quality socres (Illumina >1.3) Position in read (bp) Position in read (bp) 34 32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 38 36 34 32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 1 2 3 4 5 6 7 8 9 10-14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 20-24 30-34 40-44 50-54 60-64 70-74 80-84 90 Fig. 3 FastQC per base sequence quality boxplot example for high quality (a, sample SRR1278968) and low quality data (b, plot adapted from [29]). Each black box represents 50 % of the reads and the black lines above Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 41. 24 36 bases after trimming “-l 36”; a quality threshold for removing bases of the 3′ end with scores lower than 15 “-q 15”; a thresh- old of 15 for the mean quality of a read “-Q 15”. In the example below the output directory for the trimmed data is “trim” and “-t 4” specifies that four cores should be used when executing Skewer. The two fastq files for our sample are listed at the end of the command line, with “_1.fastq” before “_2.fastq”. mkdir trim skewer -m pe -l 36 -q 15 -Q 15 -o trim/SRR1278968 -t 4 SRR1278968_1.fastq SRR1278968_2.fastq Skewer will provide a summary for each executed command, for example for sample SRR1278968: 13722223 read pairs processed; of these: 26390 (0.19 %) short read pairs filtered out after trimming by size control 0 (0.00 %) empty read pairs filtered out after trimming by size control 13695833 (99.81 %) read pairs available; of these: 2902491 (21.19 %) trimmed read pairs available after processing 10793342 (78.81 %) untrimmed read pairs available after processing Only 0.19 % of the reads were discarded after trimming, since their length was shorter than 36 bases. From the other reads, 21.19 % were trimmed by a varying number of bases from the 3′ end because the base quality fell below the specified thresh- old of 15 (“-q 15”). 5. After trimming the data, run FastQC again, this time using the output FASTQ files from the trim folder. This ensues all the data is of high-quality (not shown). All RNA-seq reads must be mapped to a reference genome, using an aligner such as TopHat [31]. 1. Download the C. parapsilosis reference genome and gene annotation [22, 32–35] from http:/ /www.cgob.ie/supp_data. The files are called “cpar.fa” and “cpar.gff”, respectively. 2. Rename the files generated by Skewer and create an index for the reference genome. For paired-end reads, TopHat requires that the FASTQ files end in “_1.fastq” and “_2.fastq”, for the first and second mate respectively. For single-end reads no naming convention or order of samples exists. Renaming the trimmed FASTQ files that currently end with “pair1.fastq” and “pair2. fastq” is easily achieved using the Unix rename command: rename 's/pair/pair_/' *pair* 3.4.2 Mapping Reads to the Genome Can Wang et al.
- 42. 25 3. To create the reference genome index use bowtie2-build [36] with the FASTA genome file and output folder “cpar-index”: mkdir cpar-index bowtie2-build cpar.fsa cpar-index/cpar 4. Execute TopHat with the following command: tophat -p 12 -o SRR1278968 -G cpar.gff -g 1 --b2-very- sensitive --library-type fr-firststrand cpar-index/cpar trim/SRR1278968-pair_1.fastq trim/SRR1278968-pair_2. fastq The option “-p” sets the number of processing cores TopHat utilizes, “-o” sets the output folder, “-G” is optional and pro- vides genome annotation and “-g 1” sets the maximum amount of times a read can map to the genome before it is reported as ambiguously mapped. The preset “--b2-very-sensitive” is specified, which includes a number of settings (-D 20 -R 3 -N 0 -L 20 -i S,1,0.50). The D and R options specify “effort” options of TopHat. The higher these numbers are, the higher the amount of attempts TopHat will execute to realign reads or extend existing alignments. The N, L and i options fine-tune how TopHat tries to align the reads. The number of mismatches that are allowed during seed alignment (N), the length of the seed substring (L) and the function for the interval between substrings (i). Further information on TopHat can be found in the online manual (follow link in Table 4). The very-sensitive option is used to increase the probabil- ity of mapping reads, as well as the length of the alignment. If the full read does not map to the reference genome, it is cut into smaller pieces (seeds) that TopHat tries to realign. To specify that the RNA-seq data is strand-specific, set the library-type option for TopHat (“--library-type fr-firststrand”) (see Note 11). The C. parapsilosis genome, like many other eukaryotes, includes several introns [34, 35]. For this reason, a splice- aware aligner is used to map reads to the reference genome. TopHat has this ability, as do other tools (see Note 12). Aligning the reads generated several files. The most impor- tant one is “accepted_hits.bam”, which includes all reads that were successfully mapped to the genome, as well as the map- ping quality and the mapping position. The “unmapped.bam” file lists all reads that were not successfully mapped to the genome. The files “deletions.bed” and “insertions.bed” show positions where reads were successfully mapped to the genome, but compared to the reference genome bases were either miss- ing in the read (included in “deletions.bed”) or additional Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 43. 26 bases were present in the read (“insertions.bed”). The file “junctions.bed” lists all positions in the genome where reads would span a region. This includes regions where reads span an intron. Visualizing the junctions in a genome browser can help identify different isoforms of a gene. 5. To check the data for properly aligned mate pairs, generate a summary of the alignment file from TopHat using the flagstat script from SAMtools [37] with the following command: samtools flagstat accepted_hits.bam The summary lists the total number of reads with a detailed breakdown of the paired reads. An example output is shown below. Here, 97.25 % of aligned reads are properly paired and only a very minor subset of reads without a mate or with a mate mapping to a different chromosome are present. 25680571 + 0 in total (QC-passed reads + QC-failed reads) 0 + 0 duplicates 25680571 + 0 mapped (100.00%:-nan%) 25680571 + 0 paired in sequencing 12793758 + 0 read1 12886813 + 0 read2 24974160 + 0 properly paired (97.25%:-nan%) 25135208 + 0 with itself and mate mapped 545363 + 0 singletons (2.12%:-nan%) 41182 + 0 with mate mapped to a different chr 41182 + 0 with mate mapped to a different chr (mapQ>=5) 6. When the reads are mapped to the genome, prepare the aligned “.bam” files for further analysis and viewing in a genome browser. Visualizing the reads at this point in the analysis is a good way to identify any potential problems, such as incorrect mapping of paired mates, or incorrect orientation of strand- specific data. First index the “accepted_hits.bam” file. This is essential for genome browsers to display reads efficiently. The following SAMtools command generates a “.bai” file, which contains the bam index. samtools index accepted_hits.bam accepted_hits.bam.bai 1. Install the IGV browser as described in “Installing Software” (see Note 13). 2. Load the reference genome sequence and the annotation. This is achieved from a single dialog box under “Genomes -> Create .genome File…”. 3.4.3 Visualizing the Mapped Reads in a Genome Browser Can Wang et al.
- 44. 27 3. Enter a unique identifier for the genome, and select the fasta file that was used earlier for building the Bowtie2 genome index as the “FASTA file”. Alternatively, select the genome annota- tion GFF file that was used with TopHat as the “Gene file”. 4. To load BAM files into IGV, select the reference genome from the drop-down menu at the top left corner of IGV. 5. Select a BAM file generated by an aligner from the local file system, a server or URL. The accompanying BAM index file must be in the same directory as the BAM file itself. 6. Load the GFF file to display the annotation. Initially, no sequencing reads are visible. This is because the default view in IGV is to show the entire chromosome, and displaying all reads mapped to the chromosome requires too much memory. Sequencing reads will be displayed if the visible region of the chromosome is set to below 100 kb in length. A snapshot of IGV with a BAM file and genome annotation loaded is shown in Fig. 4. 7. The reads can be displayed in three different ways: collapsed, squished and expanded. This is selected by right-click on the “accepted_hits.bam” label on the left side of the track. It also can be helpful to visualize paired-end RNA-seq data. To dis- play read pairs as connected reads, select “View as pairs” again by right-click on the “accepted_hits.bam” label. By default IGV will display read pairs in different colors since the reads have different directions. To adjust the coloring schema to Fig. 4 Snapshot of IGV showing strand-specific C. parapsilosis RNA-seq data [22]. Included are RNA-seq cov- erage (top track), BAM file (middle track) and genome annotation (bottom track). Reads on the forward strand are dark grey and light grey on the reverse strand. Arrows inside the annotation track indicate the direction transcription.The data range of the RNA-seq coverage is indicated in square brackets [0-4135]. BAM file reads are displayed using the “squished” visualization option and colored using the “first-of-pair strand” option Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 45. 28 show transcriptional orientation, right-click into the “accepted_ hits.bam” label again and choose “Color alignments by” -> “first-of-pair-strand”. To measure transcripts, the number of mapped reads for each gene must be counted. The Python script htseq-count from HTSeq [38] does exactly this. However, in order to run htseq-count, mapped reads must first be sorted in the bam file according to their location on the genome, and the file converted into the non-binary and significantly larger SAM format. 1. Sort the reads by location (option “-n”) and convert the sorted BAM file to the SAM format using the following two SAMtool commands. A descriptive header for the SAM file is included with the option “-h”. samtools sort -n accepted_hits.bam accepted_hits.sorted samtools view -h -o accepted_hits.sorted.sam accepted_hits. sorted.bam 2. To run htseq-count, specify the following command for each sample: htseq-count -m union -s reverse -t exon -i transcript_id -o accepted_hits.sorted.sam.htseq accepted_hits.sorted.sam ../cpar.gff 1>accepted_hits.sorted.sam.htseq.count 2>accepted_hits.sorted.sam.htseq.count.log The option “–m union” specifies how the HTSeq algorithm assigns a read to a gene (also referred to as “feature”). The union option is recommended in most cases [38]. The strand direction (-s), the feature type (-t, third column in the GFF file), and the attribute for that htseq-count should report the read counts (-i), e.g., for each exon, or for each transcript, should also be specified. After successfully executing htseq-count the count data will be stored in “accepted_hits.sorted.sam.htseq.count”. This is a tab-delimited file with transcript names in the first column and read counts in the second: CPAR2_100010_mRNA 271 CPAR2_100020_mRNA 454 CPAR2_100030_mRNA 3277 In the following section we will explain how to identify differ- entially expressed genes from read count data and subsequently uncover the biological differences between different conditions. Identify differentially expressed genes from read count data using the Bioconductor package DESeq2, which assumes that the read count data follows a negative binomial distribution [12]. (see Note 14). 3.4.4 Counting Transcripts 3.4.5 Differential Gene Expression Analysis Can Wang et al.
- 46. 29 Run Packages in R from the R command line or from the graphical user interface. (see Note 15). 1. Once R and DESeq2 are installed (see “Installing Software” for instructions) open R and load DESeq2 and ReportingTools [39] into the R environment using the following commands: library("DESeq2") library("ReportingTools") 2. Set the working directory to the analysis folder, for example: setwd("~/ngs/data/") 3. The count data are stored inside each TopHat output folder as specified for the TopHat command in “Mapping reads to the genome”. To read these into R, use the function read.table(): samples <- c("SRR1278968","SRR1278969","SRR1278970", "SRR1278971","SRR1278972","SRR1278973") cDataAll <- NULL for(i in 1:length(samples)){ file <- read.table( sprintf("%s/accepted_hits.sorted.sam.htseq.count", samples[i])) cDataAll <- cbind(cDataAll, file[,2]) } rownames(cDataAll) <- file[,1] colnames(cDataAll) <- samples 4. The count data from all six samples is now stored in the “cDataAll” variable. The example data (Table 3) includes measurements from three planktonic and three biofilm sam- ples. To specify the conditions for each sample create the vari- able “groups” with “P” for planktonic and “B” for biofilm (see Note 16): groups <- factor(x=c(rep("P", 3), rep("B", 3)), levels=c("P", "B")) 5. The htseq-count data from Subheading “Counting transcripts” includes additional rows that indicate how many reads could not be associated with a unique feature, or where the align- ment quality was too low. Exclude these five rows from the analysis using the match() function: cData <- cDataAll[-match(x=c("__no_feature", "__ambiguous", "__too_low_aQual", "__not_aligned", "__alignment_not_ unique"), table=rownames(cDataAll)),] 6. DESeq2 uses raw count data as input, because it normalizes the data internally. To get a compact overview plot of all count Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 47. 30 data, use Tags Per Million (TPM) normalization and the density function: tpm <- t(t(cData)/colSums(cData))*1e6 inlog <- log(tpm) colLabel <- c(rep("#E41A1C", 3), rep("#377EB8", 3)) colTy <- c(rep(1:3, 3), rep(1:3, 3)) plot(density(inlog[,1]), ylim=c(0,0.4), main="Density plot of counts per gene", lty=colTy[1], xlab="Log of TPM per gene", ylab="Density", col=colLabel[1]) for(i in 2:ncol(tpm)){ lines(density(inlog[,i]), lty=colTy[i], col=colLabel[i]) } legend("topright", legend=colnames(tpm), lty=colTy, col= colLabel) This generates a density plot that shows the distribution of the log transformed TPM count data for each sample. It should follow a negative binomial distribution and with maximum log(TPM) between 4 and 5. All samples should have a very similar distribution. If this is not the case for any one sample, it is an early indicator that the transcriptome is very different to other samples. This could be due to a number of reasons. If no quality issues were detected in the raw sequencing reads and the mapping frequency to the reference genome was above 95 %, it may indicate that there is a problem with the biological sample. 7. The Principal Coordinate Analysis (PCoA) plot can also be used to characterize how similar samples are. PCoA returns coordinates that represent the dissimilarities between samples as distances. Use the normal plot() function to create the PCoA plot: d <- dist(t(tpm)) fit=cmdscale(d, eig=TRUE, k=2) x=fit$points[,1] y=fit$points[,2] plot(x, y, type="p", pch=20) text(x, y, labels=row.names(t(tpm)), cex=1, adj=c(-0.25,-0.25)) Figure 5 shows the PCoA plot using data from Table 3. Control and test samples should cluster separately and samples within each group should cluster together. The x-axis in Fig. 5 (PCoA dimension 1) clearly separates the control and test samples and the y-axis (PCoA dimension 2) indicates small differences within each group, but with a 10× smaller scale. Can Wang et al.
- 48. 31 8. When you are confident that the data does not have any bias or other confounding factors, carry out differential expression analysis. To keep structure in the analysis directory, create the folder DESeq2 and set it as the working directory. if (file.exists("DESeq2")){ setwd("DESeq2") } else { dir.create("DESeq2") setwd("DESeq2") } 9. Use the following script to run DESeq2: colData <- DataFrame(condition=groups) dds <- DESeqDataSetFromMatrix(cData, colData, formula (~condition)) dds <- DESeq(dds) res <- results(dds, cooksCutoff=FALSE) −40000 −20000 0 20000 −4000 −2000 0 2000 4000 6000 Dimension 1 Dimension 2 P3 P2 P1 B3 B2 B1 Fig. 5 PCoA plot showing transcriptional profiles of C. parapsilosis cells grown in planktonicconditions(P1-P3,SRAIDs:SRR1287968,SRR1278969,SRR1278970) and biofilm conditions (B1-B3, SRA IDs: SRR1278971, SRR1278972, SRR1278973) [22].The two groups are visually separated by Dimension 1.There is minor variation among the biological replicates, which is indicated by the ten- fold smaller scale for Dimension 2 compared to Dimension 1 Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 49. 32 The Data.Frame “colData” contains the group factor that DESeq2 uses to separate the samples. The DESeqDataSet FromMatrix() function takes the count data, sample groups, and the formula for comparing the samples as input and creates a DESeqDataSet object, which can be used to run DESeq2. The function DESeq() executes DESeq and writes all results into the DESeqDataSet object. Use the results() function to create a Data.Frame containing the results, such as gene name, log2 fold change, and adjusted p-value. 10. To write the results to a file that can be opened with Excel use the write.csv() function. In addition the result Data.Frame can be ranked by the log2 fold change to enable easier analysis of the data. res <- res[order(res$log2FoldChange, decreasing=TRUE),] write.csv(as.data.frame(res), file="p-vs-b_results.csv") Below are the first rows of the results from the CSV file. geneID baseMean log2FoldChange lfcSE stat pvalue padj CPAR2_203270_ mRNA 5478.47 11.33 0.42 26.44 3.91e−154 7.59e−152 CPAR2_807700_ mRNA 20856.65 9.68 0.17 54.60 0 0 11. To extract the numbers of significantly upregulated and down- regulated genes you must specify the results thresholds. We recommend using a log2 fold change greater than 1 for genes with increased expression or lower than −1 for genes with decreased expression. Set the significant adjusted p-value (padj) to less than 0.01. up <- rownames(res[!is.na(res$padj) & res$padj <= 0.01 & res$log2FoldChange >= 1, ]) down <- rownames(res[!is.na(res$padj) & res$padj <= 0.01 & res$log2FoldChange <= -1, ]) sprintf("%s genes up-regulated, %s genes down-regulated", length(up), length(down)) 12. To enable downstream analysis of the differentially expressed genes, save the gene IDs in two files, called “p-vs-b_up- regulated.txt” and “p-vs-b_down-regulated.txt” with the com- mand below. In the GFF file the gene IDs contain the ending “_mRNA”, which is removed using the R function sub() with the pattern “_mRNA” and an empty replacement “”. write.table(sub(pattern = "_mRNA", replacement = "", x = up), file="p-vs-b_up-regulated.txt", col.names=FALSE, row.names=FALSE, quote = FALSE) Can Wang et al.
- 50. 33 write.table(sub(pattern = "_mRNA", replacement = "", x = down), file="p-vs-b_down-regulated.txt", col.names=FALSE, row.names=FALSE, quote = FALSE) 13. The DESeq2 package contains several methods to create over- view plots of the differentially expressed genes. Use MA plot (plotMA(dds)) to create a plot of the log2 fold changes against mean normalized counts for each gene. Use the plotPCA function to create a Principal Component Analysis (PCA) plot. The number of genes that are taken into account for the dis- tance calculation of the PCA can be specified. With “ntop=500” the 500 genes with the highest row variance, i.e., the highest variance of read counts per gene across all samples, will be used for the PCA. plotPCA(dds, ntop=500) 14. To represent the log2 fold change distribution of all genes as a histogram highlighting differentially expressed genes for exam- ple in grey (as shown in Fig. 6), use the commands: hist(res[!is.na(res$padj) & res$padj <= 0.01, ][,"log2Fold Change"], breaks=seq(-15,15,0.25), col=c(rep("tomato", 56), rep("white", 8), rep("tomato", 56)), xlab="Log2 Fold Change", main="Overall Log2 fold change, p.adj<0.01") (see Note 17). Log2 Fold Change Frequency −15 −10 −5 0 5 10 15 0 50 100 150 200 250 300 350 Fig. 6 Histogram of log2 fold changes obtained by comparing the transcriptional profiles of C. parapsilosis cells grown in planktonic vs. biofilm conditions [22] using the Bioconductor package DESeq2. Significant log2 fold changes are colored in dark grey (greater than 1, or less than −1), not significant log2 fold changes are in white Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 51. 34 As a last step, searchable HTML reports of the differentially expressed genes can easily be generated using the Bioconductor package ReportingTools [39]. 1. Load ReportingTools into the R environment. library("ReportingTools") 2. Use the functions HTMLReport() and publish() to generate a website from the results data.frame created by DESeq2. htmlRep <- HTMLReport(shortName="P-vs-B_results", reportDirectory = "./reports") publish(cbind(GeneID=rownames(res),as.data.frame(res)), htmlRep) 3. Finally, create the report. finish(htmlRep) 1. After the list of significantly differentially expressed genes is generated, several tools and websites can help to identify the biological mechanisms that drive the change between the two conditions, planktonic and biofilm transcriptomes in this example. Meta-information for Candida species is available from the Candida Genome Database (CGD, [40]). Equivalent sites for other genomes include Saccharomyces species from the Saccharomyces Genome Database (SGD, [41]) and Aspergillus species from the Aspergillus Genome Database (AspGD, [42]). For less characterized species, working with homologs from a closely related species can provide more biological insight. Homology information for Candida and Saccharomyces species is available from the Candida Gene Order Browser (CGOB, [35, 43]) and the Yeast Gene Order Browser (YGOB, [44]), as well as from AspGD. 2. Gene Ontology (GO) Analysis: Using the identified differen- tially expressed genes from the C. parapsilosis planktonic and biofilm samples, Gene Ontology terms that are enriched in upregulated or downregulated genes are identified using the GO Term Finder at CGD ([40], Table 4). In step 1 on the GO Term Finder website, choose Candida parapsilosis as the target species. For step 2 upload the file “p-vs-b_up-regulated.txt” or the equivalent file containing downregulated genes, which were saved at the end of the DESeq2 workflow. Choose one of the three Gene Ontologies in step 3, i.e., Biological Process, Molecular Function or Cellular Component. We want to know if our upregulated genes have any enriched GO terms in the Molecular Function ontology. Fig. 7 shows a screenshot of the GO Term Finder with sample settings. To start the analysis, click on the Search-button in the bottom left corner. 3.4.6 Generating HTML Reports 3.4.7 Downstream Analysis of Differentially Expressed Genes Can Wang et al.
- 52. 35 Results from the GO Term Finder can be downloaded in form of an Excel file at the bottom of the page. Alternatively the results can be saved by right-clicking on the page in the browser and selecting “Save As”. This will also save the GO tree picture. In the GO tree for upregulated genes (“p-vs-b_up-regulated. txt”) Molecular Function GO terms that are significantly shared between the list of upregulated genes include Oxidoreductase Activity, Transmembrane Transporter Activity, and Transition Metal Iron Binding. For relatively little bench time, RNA-seq can yield a large amount of data. With an established protocol and workflow, RNA-seq experiments can have a quick turnaround, with the longest waiting time being the actual sequencing itself. Even commercial compa- nies are reducing this time, with some offering a turnaround of 4–6 weeks. RNA-seq holds the potential to answer key research questions. With new strategies emerging for increased multiplexing and the availability of more whole genome sequencing data and reference genomes, the uses and benefits of RNA-seq and other NGS tech- niques are becoming widespread throughout the research commu- nity. They will soon be a staple technique in the lab environment. 4 Notes 1. We recommend using a commercial kit to isolate high quality RNA. The Ribopure Yeast RNA extraction kit from Ambion is particularly useful, but other kits or methods can be used pro- viding the quality of the RNA produced is high. 2. It is now possible to design and synthesize long adapter sequences, using updated recommendations from Illumina 3.5 Final Remarks Fig. 7 Screenshot of the GO Term Finder from the Candida Genome Database. The species C. parapsilosis is selected in Step 1 and a file containing gene names is specified in Step 2. The Molecular Function Gene Ontology is selected in Step 3 Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 53. 36 [21] that remove the necessity to add the P5 region during library amplification (Fig 2b [21]). The P5 sequence is included in the first oligonucleotide (the universal adapter). The index- ing sequence is contained in the second oligonucleotide (indexed adapter), outside the short region that anneals with the universal adapter (Fig 2b). The sequence of twenty-seven 6 nucleotide indexes are provided by Illumina ([21], Oligo- nucleotide sequences, 2007–2013 Illumina, Inc. All rights reserved) and other home-made designs are described by Ford et al. [45]. Multiplexing of samples can be increased by also including an index sequence in the universal adapter (dual indexing). Recent kits from Illumina use six or eight nucleo- tide indexes, with up to eight different versions of the universal adapter, and 48 of the indexed adapter [21]. The libraries are amplified using regions derived from the P5 and P7 regions, and can be sequenced from either end using two sequencing primers. These adapters can also be combined with dUTP methodology for strand-specific sequencing [18]. 3. AMPure XP beads (Beckman Coulter) are now recommended for clean-up procedures instead of gel purification. 4. We use 2-log DNA ladder from NEB, which allows visualiza- tion of DNA bands for 0.1–10 kb. Make 2-log DNA ladder mix by mixing 1 μl 2-log DNA ladder (NEB), 1 μl Blue load- ing dye (Promega), and 4 μl distilled water. 5. Consideration should be given to the length of the reads generated, and whether single-end or paired-end sequences are used. The adapters described here are for single read only, different sequences are required for paired end reads. It is pos- sible to obtain increasingly long reads, but at a price. Long reads (>100 bases) are not necessary for RNA-seq. Paired-end reads can help in mapping, but are not strictly necessary. Strand-specific information however is strongly recommended as it enables identification of UTRs (untranslated regions) and antisense expression. 6. For users who are not yet proficient with the Terminal com- mand line, a detailed Unix tutorial is available [46]. Commands and techniques described in the tutorial are applicable for both Linux and Mac OS X users. A large community of researchers that work with next-generation sequencing data can be reached on the SEQanswers forum [47]. BioStar [47, 48] is a more general and also very useful Bioinformatics forum. 7. If any error messages arise during the installation of Python, or if there are general questions about Unix environment vari- ables or Unix commands, Stack Overflow [49] is a useful start- ing point to search for solutions. 8. More information on SRA formats are available in the SRA Knowledge Base [50] and the SRA Handbook [51]. Can Wang et al.
- 54. 37 9. More information about the GFF format is available at a dedi- cated website from the Sanger Institute [52]. 10. There are several equivalently suitable tools available for trim- ming sequencing reads, some with extensive options to specify quality thresholds and filtering parameters. It is however for the user to decide which tool fits best for a specific task. The tool used in this RNA-seq workflow is Skewer [30]. Other popular tools include fastx_trimmer [27, 28], cutadapt [53], and TrimGalore! [54]. These tools are all capable of trimming sequencing reads based on quality thresholds. 11. There are different methods for generating strand-specific RNA-seq libraries [16]. To verify that the correct option for the RNA-seq data was used in TopHat, compare the aligned reads in a genome browser with the HTSeq count data for a specific gene. For unstranded RNA-seq data specify “fr-unstranded”. 12. Other splice-aware aligners include GSNAP [55] or STAR [56]. STAR is a recently developed alignment tool that has significant speed advantages over other aligners. This is help- ful for identifying the optimal parameters for aligning sequenc- ing reads to a reference genome, e.g., number of allowed mismatches per read, number of times a read is allowed to map to the genome, or the length limitations for introns. A comparison of the different aligners was carried out by Engstrom et al. [57]. 13. Other popular genome browsers are Artemis [58], the Integrative Genomics Browser [25], the web-browser based genome browsers JBrowse [59] and the UCSC Genome Browser [60]. 14. Selecting a statistical package to identify differentially expressed genes from gene count data is an important step in an RNA- seq workflow. There is however not one method that is supe- rior to all others. Commonly used packages include DESeq2, edgeR, baySeq, DEGSeq, NOISeq, tweeDEseq, and many more. The Bioconductor version 2.14 lists 138 packages used for Differential Expression analysis. DESeq2, edgeR, baySeq, and EBSeq assume that the count data follows a negative bino- mial distribution, which is the most commonly assumed distri- bution for RNA-seq data. Choosing any of the available methods will yield results, more important than the method itself however is that experiments are planned with enough replicates and that options recommended by the authors of each method are used. It is not ideal to switch between statisti- cal methods when comparing different datasets, which could introduce additional biases. Detailed comparisons and analyses of several statistical methods for identifying differentially expressed genes from count data can be found at [61, 62]. Using RNA-seq for Analysis of Differential Gene Expression in Fungal Species
- 55. Random documents with unrelated content Scribd suggests to you:
- 56. declared that he would sooner or later have him, the Talmud, put to death by the hangman![48] For the benefit of the average reader as well as to illuminate the general subject, a short description of the Talmud will be given. Definition.—Many attempts have been made to define the Talmud, but all definition of this monumental literary production is necessarily inaccurate and incomplete because of the vastness and peculiarity of the matter treated. To describe it as an encyclopedia of the life and literature, law and religion, art and science of the Hebrew people during a thousand years would convey only an approximately correct idea of its true meaning, for it is even more than the foregoing descriptive terms would indicate. Emanuel Deutsch in his brilliant essay on the Talmud defines it as "a Corpus Juris, an encyclopedia of law, civil and penal, ecclesiastical and international, human and divine. It is a microcosm, embracing, even as does the Bible, heaven and earth. It is as if all the prose and poetry, the science, the faith and speculation of the Old World were, though only in faint reflections, bound up in it in nuce." Benny describes it as "the Talmud—that much maligned and even more misunderstood compilation of the rabbins; that digest of what Carlyle would term allerlei-wissenschaften; which is at once the compendium of their literature, the storehouse of their tradition, the exponent of their faith, the record of their acquirements, the handbook of their ceremonials and the summary of their legal code, civil and penal." To speak of the Talmud as a book would be inaccurate. It is a small library, or collection of books. "Modern editions of the Talmud, including the most important commentaries, consist of about 3,000 folio sheets, or 12,000 folio pages of closely printed matter, generally divided into twelve or twenty volumes. One page of Talmudic Hebrew intelligibly translated into English would cover three pages; the translation of the whole Talmud with its commentaries would
- 57. accordingly make a library of 400 volumes, each numbering 360 octavo pages."[49] It would be well to bear in mind that the contents of the Talmud were not proclaimed to the world by any executive, legislative, or judicial body; that they were not the result of any resolution or mandate of any congregation, college, or Sanhedrin; that they were not, in any case, formal or statutory. They were simply a great mass of traditionary matter and commentary transmitted orally through many centuries before being finally reduced to writing. Rabbinism claims for these traditions a remote antiquity, declaring them to be coeval with the proclamation of the Decalogue. Many learned doctors among the Jews ascribe this antiquity to the whole mass of traditional laws. Others maintain that only the principles upon which Rabbinic interpretation and discussion are based, can be traced back so far. But it is certain that distinct traditions are to be found at a very early period in the history of the children of Israel, and that on their return from Babylonian captivity these traditions were delivered to them by Ezra and his coadjutors of the Great Assembly. This development of Hebrew jurisprudence along lines of written and oral law, Pentateuch and Talmud, Mosaic ordinance and time- honored tradition, seems to have followed in obedience to a general principle of juristic growth. Lex scripta and lex non scripta are classical Roman terms of universal application in systems of enlightened jurisprudence. A charter, a parchment, a marble column, a table of stone, a sacred book, containing written maxims defining legal rights and wrongs are the beginnings of all civilized schemes of justice. Around these written, fundamental laws grow and cluster the race traditions of a people which attach themselves to and become inseparable from the prime organic structure. These oral traditions are the natural and necessary products of a nation's growth and progress. The laws of the Medes and Persians, at once unalterable and irrevocable, represent a strange and painful anomaly in the jurisprudence of mankind. No written constitution, incapable of amendment and subject to strict construction, can long survive the
- 58. growth and expansion of a great and progressive people. The ever- changing, perpetually evolving forms of social, commercial, political, and religious life of a restless, marching, ambitious race, necessitate corresponding changes and evolutions in laws and constitutions. These necessary legal supplements are as varied in origin as are the nations that produce them. Magna Charta, wrung from John at Runnymede, became the written basis of English law and freedom, and around it grew up those customs and traditions that—born on the shores of the German Ocean, transplanted to the Isles of Britain, nurtured and developed through a thousand years of judicial interpretation and application—became the great basic structure of the Common Law of England. What the Mosaic Code was to the ancient Hebrews, what Magna Charta is to Englishmen, the Koran is to Mahometans: the written charter of their faith and law. Surrounding the Koran are many volumes of tradition, made up of the sayings of Mahomet, which are regarded as equally sacred and authoritative as the Koran itself. These volumes of Mahometan tradition are called the Sonna and correspond to the Talmud of the Hebrews. An analysis of any great system of jurisprudence will reveal the same natural arrangement of written and oral law as that represented by the Pentateuch and the Talmud of the Jews. The word "Talmud" has various meanings, as it appears in Hebrew traditional literature. It is an old scholastic term, and "is a noun formed from the verb 'limmed'='to teach.' It therefore means, primarily, 'teaching,' although it denotes also 'learning'; it is employed in this latter sense with special reference to the Torah, the terms 'Talmud' and 'Torah' being usually combined to indicate the study of the Law, both in its wider and its more restricted sense."[50] It is thus frequently used in the sense of the word "exegesis," meaning Biblical exposition or interpretation. But with the etymological and restricted, we are not so much interested as with the popular and general signification of the term "Talmud." Popularly used, it means simply a small collection of books represented by two
- 59. distinct editions handed down to posterity by the Palestinian and Babylonian schools during the early centuries of the Christian era. Divisions of the Talmud.—The Talmud is divided into two component parts: the Mishna, which may be described as the text; and the Gemara, which may be termed the commentary.[51] The Mishna, meaning tradition, is almost wholly law. It was, indeed, of old, translated as the Second or Oral Law—the δευτέρωσιϛ—to distinguish it from the Written Law delivered by God to Moses. The relationship between the Mishna, meaning oral law, and the Gemara, meaning commentary, may be illustrated by a bill introduced into Congress and the debates which follow. In a general way, the bill corresponds to the Mishna, and the debates to the Gemara. The distinction, however, is that the law resulting from the passage of the bill is the effect and culmination of the debate; while the Mishna was already law when the Gemara or commentary was made. As we have seen above, Hebrew jurisprudence in its principles and in the manner of their interpretation was chiefly transmitted by the living voice of tradition. These laws were easily and safely handed down from father to son through successive generations as long as Jewish nationality continued and the Temple at Jerusalem still stood. But, with the destruction of the Temple and the banishment of the Jews from Palestine (A.D. 70), the danger became imminent that in the loss of their nationality would also be buried the remembrance of their laws. Moved with pity and compassion for the sad condition of his people, Judah the Holy, called Rabbi for preëminence, resolved to collect and perpetuate for them in writing their time-honored traditions. His work received the name Mishna, the same which we have discussed above. But it must not be imagined that this work was the sudden or exclusive effort of Rabbi Judah. His achievement was merely the sum total and culmination of the labors of a long line of celebrated Hebrew sages. "The Oral Law had been recognized by Ezra; had become important in the days of the Maccabees; had been supported by Pharisaism; narrowed by the school of Shammai, codified by the school of Hillel, systematized by R. Akiba, placed on
- 60. a logical basis by R. Ishmael, exegetically amplified by R. Eliezer, and constantly enriched by successive rabbis and their schools. Rabbi Judah put the coping-stone to the immense structure."[52] Emanuel Deutsch gives the following subdivisions of the Mishna: The Mishna is divided into six sections. These are subdivided again into 11, 12, 7, 9 (or 10), 11, and 12 chapters, respectively, which are further broken up into 524 paragraphs. We shall briefly describe their contents: Section I. Seeds: of Agrarian Laws, commencing with a chapter on Prayers. In this section, the various tithes and donations due to the Priests, the Levites, and the poor, from the products of the lands, and further the Sabbatical year and the prohibited mixtures in plants, animals, garments, are treated of. Section II. Feasts: of Sabbaths, Feast, and Fast days, the work prohibited, the ceremonies ordained, the sacrifices to be offered, on them. Special chapters are devoted to the Feast of the Exodus from Egypt, to the New Year's Day, to the Day of Atonement (one of the most impressive portions of the whole book), to the Feast of Tabernacles and to that of Haman. Section III. Women: of betrothal, marriage, divorce, etc., also of vows. Section IV. Damages: including a great part of the civil and criminal law. It treats of the law of trover, of buying and selling, and the ordinary monetary transactions. Further, of the greatest crime known to the law, viz., idolatry. Next of witnesses, of oaths, of legal punishments, and of the Sanhedrin itself. This section concludes with the so-called "Sentences of the Fathers," containing some of the sublimest ethical dicta known in the history of religious philosophy.
- 61. Section V. Sacred Things: of sacrifices, the first-born, etc.; also of the measurements of the Temple (Middoth). Section VI. Purifications: of the various levitical and other hygienic laws, of impure things and persons, their purification, etc.[53] Recensions.—The Talmud exists in two recensions: the Jerusalem and the Babylonian. These two editions represent a double Gemara; the first (Jerusalem) being an expression of the schools in Palestine and redacted at Tiberias about 390 A.D.; the second (Babylonian) being an expression of the schools in Babylonia and redacted about 365-427 A.D. The Mishna, having been formed into a code, became in its turn what the Pentateuch had been before it, a basis of discussion and development. The Gemara of the Jerusalem Talmud embodies the critical discussions and disquisitions on the Mishna by hundreds of learned doctors who lived in Palestine, chiefly in Galilee, from the end of the second till about the middle of the fifth century of the Christian era. The Gemara of the Babylonian Talmud embodies the criticisms and dissertations on the same Mishna of numerous learned doctors living in various places in Babylonia, but chiefly those of the two great schools of Sura and Pumbaditha.[54] The Babylonian Talmud is written in "West Aramæan," is the product of six or seven generations of constant development, and is about four times as large as that of the Jerusalem Talmud, which is written in "East Aramæan."[55] It should be kept clearly before the mind that the only difference between these two recensions is in the matter of commentary. The two sets of doctors whose different commentaries distinguish the two Talmuds dealt with the same Mishna as a basis of criticism. But decided differences are noticeable in the subject matter and style of the two Gemaras represented by the two recensions of the Talmud. The discussions and commentaries in the Jerusalem Talmud are simple, brief, and pointed; while those of the Babylonian Talmud are generally subtle, abstruse, and prolix. The dissertations in the Jerusalem Talmud are filled to overflowing with
- 62. archæology, geography, and history, while the Babylonian Talmud is more marked by legal and religious development. But the reader should not form a wrong impression of the contents of the Talmud. They are a blending of the oral law of the Mishna and the notes and comments of the sages. The characteristics of both the editions are legal and religious, but a multitude of references are made in each to things that have no connection with either religion or law. "The Talmud does, indeed, offer us a perfect picture of the cosmopolitanism and luxury of those final days of Rome, such as but few classical or postclassical writings contain. We find mention made of Spanish fish, of Cretan apples, Bithynian cheese, Egyptian lentils and beans, Greek and Egyptian pumpkins, Italian wine, Median beer, Egyptian Zyphus; garments imported from Pelusium and India, shirts from Cilicia, and veils from Arabia. To the Arabic, Persian, and Indian materials contained, in addition to these, in the Gemara, a bare allusion may suffice. So much we venture to predict, that when once archæological and linguistic science shall turn to this field, they will not leave it again soon." Relation of Talmud to Mishna.—The relation of the Talmud, used in the popular sense, to the Mishna, raises the question of the relation of the whole to one of its parts. The varying meanings of Mishna, Gemara, and Talmud very easily confuse the ordinary reader. If these terms are considered separately in the order in which they appear in the preceding sentence, simple mathematical addition will greatly aid in elucidating matters. The Mishna is a vast mass of tradition or oral law which was finally reduced to writing about the close of the second century of the Christian era. The Gemara is the Rabbinical exposition of the meaning of the Mishna. The Talmud is the sum of the Mishna plus the Gemara. In other words, the Talmud is the elaboration or amplification of the Mishna by manifold commentaries, designated as the Gemara. It frequently happens that the Talmud and the Mishna appear in the same sentence as terms designating entirely different things. This association in a different sense inevitably breeds confusion, unless we pause to
- 63. consider that the Mishna has a separate existence from the Talmud and a distinct recension of its own. In this state it is simply a naked code of laws. But when the Gemara has been added to it the Talmud is the result, which, in its turn, becomes a distinct entity and may be referred to as such in the same sentence with the Mishna. Relation of Talmud to Pentateuch.—As before suggested, the Pentateuch, or Mosaic Code, was the Written Law and the very foundation of ancient Hebrew jurisprudence. The Talmud, composed of the Mishna, i.e., Tradition, and the Gemara, i.e., Commentary, was the Oral Law, connected with, derived from, and built upon the Written Law. It must be remembered that the commonwealth of the Jews was a pure theocracy and that all law as well as all religion emanated directly or indirectly from Jehovah. This was as true of Talmudic tradition as of Mosaic ordinance. Hillel, who interpreted tradition, was as much inspired of God as was Moses when he received the Written Law on Sinai. Emanuel Deutsch is of the opinion that from the very beginning of the Mosaic law there must have existed a number of corollary laws which were used to interpret and explain the written rules; that, besides, there were certain enactments of the primitive Council of the Desert, and certain verdicts issued by the later "judges within the gates"—all of which entered into the general body of the Oral Law and were transmitted side by side with the Written Law through the ages.[56] The fourth book of Ezra, as well as other Apocryphal writings, together with Philo and certain of the Church Fathers, tells us of great numbers of books that were given to Moses at the same time that he received the Pentateuch. These writings are doubtless the source of the popular belief among the Jews that the traditional laws of the Mishna had existed from time immemorial and were of divine origin. "Jewish tradition traces the bulk of the oral injunctions, through a chain of distinctly named authorities, to 'Sinai itself.' It mentions in detail how Moses communicated those minutiæ of his legislation, in which he had been instructed during the mysterious forty days and nights on the Mount, to the chosen guides of the people, in such a manner that they should forever remain engraven on the tablets of
- 64. their hearts."[57] This direct descent of the Oral Law from the Sacred Mount itself would indicate an independent character and authority. Nevertheless, Talmudic interpretation of tradition professed to remain always subject to the Mosaic Code; to be built upon, and to derive its highest inspiration from it. But, as a matter of fact, while claiming theoretically to be subordinate to it, the Talmud finally superseded and virtually displaced the Pentateuch as a legal and administrative code. This was the inevitable consequence and effect of the laws of growth and progress in national existence. Altered conditions of life, at home and in exile, necessitated new rules of action in the government of the Jewish commonwealth. The Mosaic Code was found inadequate to the ever-changing exigencies of Hebrew life. As a matter of fact, Moses laid down only general principles for the guidance of Hebrew judges. He furnished the body of the law, but a system of legal procedure was wholly wanting. The Talmud supplied the deficiency and completed a perfect whole. While yet in the Wilderness, Moses commanded the Israelites to establish courts and appoint judges for the administration of justice as soon as they were settled in Palestine.[58] This clearly indicates that the great lawgiver did not intend his ordinances and injunctions to be final and exclusive. Having furnished a foundation for the scheme, he anticipated that the piety, judgment, and learning of subsequent ages would do the rest. His expectations were fulfilled in the development of the traditions afterwards embodied in the Mishna, which is the principal component part of the Talmud. As before suggested, with the growth in population and the ever- increasing complications in social, political, and religious life, and with the general advance in Hebrew civilization, Mosaic injunction began to prove entirely inadequate to the national wants. In the time intervening between the destruction of the first and second Temples, a number of Mosaic laws had become utter anachronisms; others were perfectly impracticable, and several were no longer even understood. The exigencies of an altered mode of life and the changed conditions and circumstances of the people rendered imperative the enactment of new laws unknown to the Pentateuch.
- 65. But the divine origin of the Hebrew system of law was never for a moment forgotten, whatever the change and wherever made. The Rabbins never formally repealed or abolished any Mosaic enactment. They simply declared that it had fallen into desuetude. And, in devising new laws rendered necessary by changed conditions of life they invariably invoked some principle or interpretation of the Written Law. In the declining years of Jewish nationality, many characteristic laws of the Pentateuch had become obsolete. The ordinance which determined the punishment of a stubborn and rebellious son; the enactment which commanded the destruction of a city given to idolatry; and, above all, the lex talionis had become purely matters of legend. On the other hand, many new laws appear in the Talmud of which no trace whatever can be discovered in the Pentateuch. "The Pharisees," says Josephus, "have imposed upon the people many laws taken from the tradition of the Fathers, which are not written in the law of Moses."[59] The most significant of these is the one providing for Antecedent Warning in criminal prosecutions, the meaning and purpose of which will be fully discussed in another chapter. Vicissitudes of the Talmud.—An old Latin adage runs: "Habent sua fata libelli."[60] (Even books are victims of fate). This saying is peculiarly applicable to the Talmud, which has had, in a general way, the same fateful history as the race that created it. Proscription, exile, imprisonment, confiscation, and burning was its lot throughout the Middle Ages. During a thousand years, popes and kings vied with each other in pronouncing edicts and hurling anathemas against it. During the latter half of the sixteenth century it was burned not fewer than six different times by royal or papal decree. Whole wagonloads were consigned to the flames at a single burning. In 1286, in a letter to the Archbishop of Canterbury, Honorius IV described the Talmud as a "damnable book" (liber damnabilis), and vehemently urged that nobody in England be permitted to read it, since "all other evils flow out of it."[61] On New Year's day, 1553,
- 66. numerous copies of the Talmud were burned at Rome in compliance with a decree of the Inquisition. And, as late as 1757, in Poland, Bishop Dembowski, at the instigation of the Frankists, convened a public assembly at Kamenetz-Podolsk, which decreed that all copies of the Talmud found in the bishopric should be confiscated and burned by the hangman.[62] Of the two recensions, the Babylonian Talmud bore the brunt of persecution during all the ages. This resulted from the fact that the Jerusalem Talmud was little read after the closing of the Jewish academies in Palestine, while the Babylonian Talmud was the popular edition of eminent Jewish scholars throughout the world. It is needless to say that the treatment accorded the venerable literary compilation was due to bitter prejudice and crass ignorance. This is well illustrated by the circumstance that when, in 1307, Clement V was asked to issue a bull against the Talmud, he declined to do so, until he had learned something about it. To his amazement and chagrin, he could find no one who could throw any light upon the subject. Those who wished it condemned and burned were totally ignorant of its meaning and contents. The surprise and disgust of Clement were so great that he resolved to found three chairs in Hebrew, Arabic, and Chaldee, the three tongues nearest the idiom of the Talmud. He designated the Universities of Paris, Salamanca, Bologna, and Oxford as places where these languages should be taught, and expressed the hope that, in time, one of these universities might be able to produce a translation of "this mysterious book."[63] It may be added that these plans of the Pope were never consummated. The Message and Mission of the Talmud.—To appreciate the message and mission of the Talmud, its contents must be viewed and contemplated in the light of both literature and history. As a literary production it is a masterpiece—strange, weird, and unique— but a masterpiece, nevertheless. It is a sort of spiritual and intellectual cosmos in which the brain growth and soul burst of a great race found expression during a thousand years. As an
- 67. encyclopedia of faith and scholarship it reveals the noblest thoughts and highest aspirations of a divinely commissioned race. Whatever the master spirits of Judaism in Palestine and Babylon esteemed worthy of thought and devotion was devoted to its pages. It thus became a great twin messenger, with the Bible, of Hebrew civilization to all the races of mankind and to all the centuries yet to come. To Hebrews it is still the great storehouse of information touching the legal, political, and religious traditions of their fathers in many lands and ages. To the Biblical critic of any faith it is an invaluable help to Bible exegesis. And to all the world who care for the sacred and the solemn it is a priceless literary treasure. As an historical factor the Talmud has only remotely affected the great currents of Gentile history. But to Judaism it has been the cementing bond in every time of persecution and threatened dissolution. It was carried from Babylon to Egypt, northern Africa, Spain, Italy, France, Germany, and Poland. And when threatened with national and race destruction, the children of Abraham in every land bowed themselves above its sacred pages and caught therefrom inspiration to renewed life and higher effort. The Hebrews of every age have held the Talmud in extravagant reverence as the greatest sacred heirloom of their race. Their supreme affection for it has placed it above even the Bible. It is an adage with them that, "The Bible is salt, the Mischna pepper, the Gemara balmy spice," and Rabbi Solomon ben Joseph sings: "The Kabbala and Talmud hoar Than all the Prophets prize I more; For water is all Bible lore, But Mischna is pure wine." More than any other human agency has the Talmud been instrumental in creating that strangest of all political phenomena—a nation without a country, a race without a fatherland.
- 69. CHAPTER II HEBREW CRIMINAL LAW—CRIMES AND PUNISHMENTS apital crimes, under Hebrew law, were classified by Maimonides according to their respective penalties. His arrangement will be followed in this chapter.[64] Hebrew jurisprudence provided four methods of capital punishment: (1) Beheading; (2) Strangling; (3) Burning; (4) Stoning. Crucifixion was unknown to Hebrew law. This cruel and loathsome form of punishment will be fully discussed in the second volume of this work. Thirty-six capital crimes are mentioned by the Pentateuch and the Talmud. Beheading was the punishment for only two crimes: (1) Murder. (2) Communal apostasy from Judaism to idolatry. Strangling was prescribed for six offenses: (1) Adultery. (2) Kidnaping. (3) False prophecy. (4) Bruising a parent. (5) Prophesying in the name of heathen deities.
- 70. (6) Maladministration (the "Rebellious Elder"). Burning was the death penalty for ten forms of incest—criminal commerce: (1) With one's own daughter. (2) With one's own son's daughter. (3) With one's own daughter's daughter. (4) With one's own stepdaughter. (5) With one's own stepson's daughter. (6) With one's own stepdaughter's daughter. (7) With one's own mother-in-law. (8) With one's own mother-in-law's mother. (9) With one's own father-in-law's mother. (10) With a priest's daughter.[65] Stoning was the penalty for eighteen capital offenses: (1) Magic. (2) Idolatry. (3) Blasphemy. (4) Pythonism. (5) Pederasty. (6) Necromancy. (7) Cursing a parent. (8) Violating the Sabbath. (9) Bestiality, practiced by a man. (10) Bestiality, practiced by a woman. (11) Sacrificing one's own children to Moloch. (12) Instigating individuals to embrace idolatry. (13) Instigating communities to embrace idolatry. (14) Criminal conversation with one's own mother.
- 71. (15) Criminal conversation with a betrothed virgin. (16) Criminal conversation with one's own stepmother. (17) Criminal conversation with one's own daughter-in-law. (18) Violation of filial duty (making the "Prodigal Son").[66] The crime of false swearing requires special notice. This offense could not be classified under any of the above subdivisions because of its peculiar nature. The Mosaic Code ordains in Deut. xix. 16-21: "If a false witness rise up against any man to testify against him that which is wrong ... and, behold, if the witness be a false witness, and hath testified falsely against his brother; then shall ye do unto him, as he had thought to have done unto his brother ... and thine eye shall not pity, but life shall go for life, eye for eye, tooth for tooth, hand for hand, foot for foot." Talmudic construction of this law awarded the same kind of death to him who had sworn falsely against his brother that would have been meted out to the alleged criminal, if the testimony of the false swearer had been true. Imprisonment, as a method of punishment, was unknown to the Mosaic Code. Leviticus xxiv. 12 and Numbers xv. 34 seem to indicate the contrary; but the imprisonment therein mentioned undoubtedly refers to the mere detention of the prisoner until sentence could be pronounced against him. Imprisonment as a form of punishment was a creation of the Talmudists who legalized its application among the Hebrews. According to Mendelsohn, five different classes of offenders were punished by imprisonment: (1) Homicides; whose crime could not be legally punished with death, because some condition or other, necessary to produce a legal conviction, had not been complied with. (2) Instigators to or procurers of murder; such, for instance, as had the deed committed by the hands of a hireling. (3) Accessories to loss of life, as, for instance, when several persons had clubbed one to death, and the court could not determine the one who gave the death blow.
- 72. (4) Persons who having been twice duly condemned to and punished with flagellation for as many transgressions of one and the same negative precept, committed it a third time. (5) Incorrigible offenders, who, on each of three occasions, had failed to acknowledge as many warnings antecedent to the commission of one and the same crime, the original penalty for which was excision.[67] Flagellation is the only corporal punishment mentioned by the Pentateuch. The number of stripes administered were not to exceed forty and were to be imposed in the presence of the judges.[68] Wherever the Mosaic Code forbade an act, or, in the language of the sages, said "Thou shalt not," and prescribed no other punishment or alternative, a Court of Three might impose stripes as the penalty for wrongdoing. Mendelsohn gives the following classification: Flagellation is the penalty of three classes of offenses: (1) The violation of a negative precept, deadly in the sight of heaven. (2) The violation of any negative precept, when accomplished by means of a positive act. (3) The violation of any one of the prohibitive ordinances punishable, according to the Mosaic law with excision, to which, however, no capital punishment at the instance of a human tribunal is attached. [69] The Mishna enumerates fifty offenses punishable by stripes, but this enumeration is evidently incomplete. Maimonides gives a full classification of all the offenses punishable by flagellation, the number of which he estimates to be two hundred and seven. The last three in his list are cases in which the king takes too many wives, accumulates too much silver or gold, or collects too many horses.[70]
- 73. Slavery was the penalty for theft under ancient Hebrew law. This is the only case where the Mosaic law imposed slavery upon the culprit as a punishment for his crime; and a loss of liberty followed only where the thief was unable to make the prescribed restitution. Exodus xxii. 1-3 says: If a man shall steal an ox, or a sheep, and kill it, or sell it, he shall restore five oxen for an ox, and four sheep for a sheep ... if he have nothing, then he shall be sold for his theft. Penal servitude, or slavery, was imposed only on men, never on women. Slavery, as a penalty for theft, was limited to a period of six years in obedience to the Mosaic ordinance laid down in Exodus xxi. 2. If thou buy a Hebrew servant, six years he shall serve: and in the seventh, he shall go free for nothing. It should be remarked, in this connection, that slavery, as a punishment for crime, carried with it none of the odium and hardship usually borne by the slave. The humanity of Hebrew law provided that the culprit, thief though he was, should not be degraded or humiliated. He could be compelled to do work for his master, such as he had been accustomed to do while free, but was relieved by the law from all degrading employment, such as "attending the master to the bath, fastening or unfastening his sandals, washing his feet, or any other labor usually performed by the regular slave." Hebrew law required such kindly treatment of the convict thief by his master that this maxim was the result: "He who buys a Hebrew slave, buys himself a master." Internment in a city of refuge was the punishment for accidental homicide. Mischance or misadventure, resulting in the slaying of a fellow-man, was not, properly speaking, a crime; nor was exile in a city of refuge considered by the Talmudists a form of punishment. But they are so classified by most writers on Hebrew criminal law. Among nearly all ancient nations there was a place of refuge for the
- 74. unfortunate and downtrodden of the earth; debtors, slaves, criminals, and political offenders; some sacred spot—an altar, a grave, or a sanctuary dedicated and devoted to some divinity who threw about the hallowed place divine protection and inviolability. Such was at Athens the Temple of Theseus, the sanctuary of slaves. It will be remembered that the orator Demosthenes took refuge in the Temple of Poseidon as a sanctuary, when pursued by emissaries of Antipater and the Macedonians.[71] Among the ancient Hebrews, there were six cities of refuge; three on either side of the Jordan. They were so located as to be nearly opposite each other. Bezer in Reuben was opposite Hebron in Judah; Schechem in Ephraim was opposite to Ramoth in Gad; and Golan in Manasseh was opposite to Kedesh in Naphtali.[72] Highways in excellent condition led from one to the other. Signposts were placed at regular intervals to indicate the way to the nearest city of refuge. These cities were designated by the law as asylums or sanctuaries for the protection of innocent slayers of their fellow-men from the "avenger of blood." Among nearly all primitive peoples of crude political development, such as the early Germans, the ancient Greeks and Slavs, certain North American savage tribes and the modern Arabs, Corsicans and Sicilians, the right of private vengeance was and is taught and tolerated. Upon the "next of kin," the "avenger of blood," devolved the duty of hunting down and slaying the guilty man. Cities of refuge were provided by Mosaic law for such an emergency among the Hebrews. This provision of the Mosaic Code doubtless sprang from a personal experience of its founder. Bible students will remember that Moses slew an Egyptian and was compelled to flee in consequence. [73] Remembering his dire distress on this occasion, the great lawgiver was naturally disposed to provide sanctuaries for others similarly distressed. But the popular notion of the rights of sanctuary under the Mosaic law is far from right. That a common murderer could, by precipitate flight, reach one of the designated places and be safe from his pursuers and the vengeance of the law, is thought by many. The observation of Benny on this point is apt and lucid:
- 75. Internment in one of the cities of refuge was not the scampering process depicted in the popular engraving: a man in the last stage of exhaustion at the gate of an Eastern town; his pursuers close upon him, arrows fixed and bows drawn; his arms stretched imploringly towards a fair Jewish damsel, with a pitcher gracefully poised upon her head. This may be extremely picturesque, but it is miserably unlike the custom in vogue among the later Hebrews. Internment in a city of refuge was a sober and judicial proceeding. He who claimed the privilege was tried before the Sanhedrin like any ordinary criminal. He was required to undergo examination; to confront witnesses, to produce evidence, precisely as in the case of other offenders. He had to prove that the homicide was purely accidental; that he had borne no malice against his neighbor; that he had not lain in wait for him to slay him. Only when the judges were convinced that the crime was homicide by misadventure was the culprit adjudged to be interned in one of the sheltering cities. There was no scurrying in the matter; no abrupt flight; no hot pursuit, and no appeal for shelter. As soon as judgment was pronounced the criminal was conducted to one of the appointed places. He was accompanied the whole distance by two talmide-chachamin-disciples of the Rabbins. The avengers of the blood dared not interfere with the offender on the way. To slay him would have been murder, punishable with death. Execution of Capital Sentences. (1) Beheading.—The Hebrews considered beheading the most awful and ignominious of all forms of punishment. It was the penalty for deliberate murder and for communal apostasy from Judaism to idolatry, the most heinous offenses against the Hebrew theocracy. Beheading was accomplished by fastening the culprit securely to a post and then severing his head from his body by a stroke with a sword.[74]
- 76. (2) Strangling.—The capital punishment of strangling was effected by burying the culprit to his waist in soft mud, and then tightening a cord wrapped in a soft cloth around his neck, until suffocation ensued.[75] (3) Burning.—The execution of criminals by burning was not done by consuming the living person with fire, as was practiced in the case of heretics by prelates in the Middle Ages and in the case of white captives by savages in colonial days in America. Indeed, the term "burning" seems to be a misnomer in this connection, for the culprit was not really burned to death. He was simply suffocated by strangling. As in the case of strangling, the condemned man was placed in a pit dug in the ground. Soft dirt was then thrown in and battered down, until nothing but his head and chest protruded. A cord, wrapped in a soft cloth, was then passed once around his neck. Two strong men came forward, grasped each an end, and drew the cord so hard that suffocation immediately followed. As the lower jaw dropped from insensibility and relaxation, a lighted wick was quickly thrown into his mouth. This constituted the burning.[76] There is authority for the statement that instead of a lighted wick, molten lead was poured down the culprit's throat.[77] (4) Stoning.—Death by stoning was accomplished in the following manner: The culprit was taken to some lofty hill or eminence, made to undress completely, if a man, and was then precipitated violently to the ground beneath. The fall usually broke the neck or dislocated the spinal cord. If death did not follow instantaneously the witnesses hurled upon his prostrate body heavy stones until he was dead. If the first stone, so heavy as to require two persons to carry it, did not produce death, then bystanders threw stones upon him until death ensued. Here, again, "stoning" to death is not strictly accurate. Death usually resulted from the fall of the man from the platform, scaffold, hill, or other elevation from which he was hurled. It was really a process of neck-breaking, instead of stoning, as burning was a process of suffocation, instead of consuming with fire.
- 77. These four methods of execution—beheading, strangling, burning, and stoning—were the only forms of capital punishment known to the ancient Hebrews. Crucifixion was never practiced by them; but a posthumous indignity, resembling crucifixion, was employed as an insult to the criminal, in the crimes of idolatry and blasphemy. In addition to being stoned to death, as a punishment for either of these crimes, the dead body of the culprit was then hanged in public view as a means of rendering the offense more hideous and the death more ignominious. This hanging to a tree was in obedience to a Mosaic ordinance contained in Deut. xxi. 22. The corpse was not permitted, however, to remain hanging during the night. The burial of the dead body of the criminal immediately followed execution, but interment could not take place in the family burial ground. Near each town in ancient Palestine were two cemeteries; in one of them were buried those criminals who had been executed by beheading or strangling; in the other were interred those who had been put to death by stoning or burning. The bodies were required to remain, thus buried, until the flesh had completely decayed and fallen from the bone. The relatives were then permitted to dig up the skeletons and place them in the family sepulchers.
- 78. CHAPTER III HEBREW CRIMINAL LAW—COURTS AND JUDGES HE Hebrew tribunals were three in kind: the Great Sanhedrin; the Minor Sanhedrin; and the Lower Tribunal, or the Court of Three. The Great Sanhedrin, or Grand Council, was the high court of justice and the supreme tribunal of the Jews. It sat at Jerusalem. It numbered seventy-one members. Its powers were legislative, executive, and judicial. It exercised all the functions of education, of government, and of religion. It was the national parliament of the Hebrew Theocracy, the human administrator of the divine will. It was the most august tribunal that ever interpreted or administered religion to man. The Name.—The word "Sanhedrin" is derived from the Greek (συνέδριον) and denotes a legislative assembly or an ecclesiastical council deliberating in a sitting posture. It suggests also the gravity and solemnity of an Oriental synod, transacting business of great importance. The etymology of the word indicates that it was first used in the later years of Jewish nationality. Several other names are also found in history to designate the Great Sanhedrin of the Jews. The Council of Ancients is a familiar designation of early Jewish writers. It is called Gerusia, or Senate, in the second book of Maccabees.[78] Concilium, or Grand Council, is the name found in the Vulgate.[79] The Talmud designates it sometimes as the Tribunal of the Maccabees, but usually terms it Sanhedrin, the name most frequently employed in the Greek text of the Gospels, in the writings of the Rabbins, and in the works of Josephus.[80]
- 79. Origin of the Great Sanhedrin.—The historians are at loggerheads as to the origin of the Great Sanhedrin. Many contend that it was established in the Wilderness by Moses, who acted under divine commission recorded in Numbers xi. 16, 17: "Gather unto me seventy of the elders of Israel, whom thou knowest to be the elders of the people, and officers of them; and bring them unto the tabernacle of the congregation, that they may stand with thee; and I will take of the Spirit that is upon thee and will put it upon them; and they shall bear the burden of the people with thee, that thou bearest it not alone." Over the seventy elders, Moses is said to have presided, making seventy-one, the historic number of the Great Sanhedrin. Several Christian historians, among them Grotius and Selden, have entertained this view; others equally celebrated have maintained contrary opinions. These latter contend that the council of seventy ordained by Moses existed only a short time, having been established to assist the great lawgiver in the administration of justice; and that, upon the entrance of the children of Israel into the Promised Land, it disappeared altogether. The writers who hold this view contend that if the great assembly organized in the Wilderness was perpetuated side by side with the royal power, throughout the ages, as the Rabbis maintained, some mention of this fact would, in reason, have been made by the Bible, Josephus, or Philo. The pages of Jewish history disclose the greatest diversity of opinion as to the origin of the Great Sanhedrin. The Maccabean era is thought by some to be the time of its first appearance. Others contend that the reign of John Hyrcanus, and still others that the days of Judas Maccabeus, marked its birth and beginning. Raphall, having studied with care its origin and progress, wrote: "We have thus traced the existence of a council of Zekenim or Elders founded by Moses, existing in the days of Ezekiel, restored under the name of Sabay Yehoudai, or Elders of the Jews, under Persian dominion; Gerusia, under the supremacy of the Greeks; and Sanhedrin under the Asmonean kings and under the Romans."[81]
- 80. Brushing aside mere theory and speculation, one historical fact is clear and uncontradicted, that the first Sanhedrin Council clothed with the general judicial and religious attributes of the Great Sanhedrin of the times of Jesus, was established at Jerusalem between 170 and 106 B.C. Organization of the Great Sanhedrin.—The seventy-one members composing the Great Sanhedrin were divided into three chambers: The chamber of priests; The chamber of scribes; The chamber of elders. The first of these orders represented the religious or sacerdotal; the second, the literary or legal; the third, the patriarchal, the democratic or popular element of the Hebrew population. Thus the principal Estates of the Commonwealth of Israel were present, by representation, in the great court and parliament of the nation. Matthew refers to these three orders and identifies the tribunal that passed judgment upon Christ: "From that time forth, began Jesus to shew unto his disciples, how that he must go unto Jerusalem, and suffer many things of the elders and chief priests and scribes, and be killed and raised again the third day."[82] Theoretically, under the Hebrew constitution, the "seventy-one" of the three chambers were to be equally divided: Twenty-three in the chamber of priests, Twenty-three in the chamber of scribes, Twenty-three in the chamber of elders. A total of sixty-nine, together with the two presiding officers, would constitute the requisite number, seventy-one. But, practically, this arrangement was rarely ever observed. The theocratic structure of the government of Israel and the pious regard of the people for the
- 81. guardians of the Temple, gave the priestly element a predominating influence from time to time. The scribes, too, were a most vigorous and aggressive sect and frequently encroached upon the rights and privileges of the other orders. Abarbanel, one of the greatest of the Hebrew writers, has offered this explanation: "The priests and scribes naturally predominated in the Sanhedrin because, not having like the other Israelites received lands to cultivate and improve, they had abundant time to consecrate to the study of law and justice, and thus became better qualified to act as judges."[83] Qualifications of Members of the Great Sanhedrin.—The following qualifications were requisite to entitle an applicant to membership in the Great Sanhedrin: (1) He must have been a Hebrew and a lineal descendant of Hebrew parents.[84] (2) He must have been "learned in the law"; both written and unwritten. His legal attainment must have included an intimate acquaintance with all the enactments of the Mosaic Code, with traditional practices, with the precepts and precedents of the colleges, with the adjudications of former courts and the opinions of former judges. He must have been familiar not only with the laws then actively in force, but also with those that had become obsolete.[85] (3) He must have had judicial experience; that is, he must have already filled three offices of gradually increasing dignity, beginning with one of the local courts, and passing successively through two magistracies at Jerusalem.[86] (4) He must have been thoroughly proficient in scientific knowledge. The ancient Sanhedrists were required to be especially well grounded in astronomy and medicine. They were also expected to be familiar with the arts of the necromancer.[87] We are also led to believe from the revelations of the Talmud that the judges of Israel
- 82. were well versed in the principles of physiology and chemistry, as far as these sciences were developed and understood in those days. History records that Rabbi Ismael and his disciples once engaged in experimental dissection in order to learn the anatomy of the human frame. On one occasion a deceitful witness tried to impose upon a Hebrew court by representing spermatic fluid to be the albumen of an egg. Baba bar Boutah was enabled, from his knowledge of the elements of chemistry, to demonstrate the fact of fraud in the testimony of the witness. Eighty disciples of the famous Academy of Hillel are said to have been acquainted with every branch of science known in those days.[88] (5) He must have been an accomplished linguist; that is, he must have been thoroughly familiar with the languages of the surrounding nations. Interpreters were not allowed in Hebrew courts. A knowledge of several languages was, therefore, indispensable to the candidate who sought membership in the Great Sanhedrin. "In the case of a foreigner being called as a witness before a tribunal, it was absolutely necessary that two members should understand the language in which the stranger's evidence was given; that two others should speak to him; while another was required to be both able to understand and to converse with the witness. A majority of three judges could always be obtained on any doubtful point in the interpretation of the testimony submitted to the court. At Bither there were three Rabbins acquainted with every language then known, while at Jabneh there were said to be four similarly endowed with the gift of 'all the tongues.'"[89] (6) He must have been modest, popular, of good appearance, and free from haughtiness.[90] The Hebrew mind conceived modesty to be the natural result of that learning, dignity, and piety which every judge was supposed to possess. The qualification of "popularity" did not convey the notion of electioneering, hobnobbing and familiarity. It meant simply that
- 83. the reputation of the applicant for judicial honors was so far above reproach that his countrymen could and would willingly commit all their interests of life, liberty, and property to his keeping. By "good appearance" was meant that freedom from physical blemishes and defects, and that possession of physical endowments that would inspire respect and reverence in the beholder. The haughty judge was supposed to be lacking in the elements of piety and humility which qualified him for communion with God. Haughtiness, therefore, disqualified for admission to the Great Sanhedrin. (7) He must have been pious, strong, and courageous.[91] Piety was the preëminent qualification of a judge of Israel. Impiety was the negation of everything Israelitish. Strength and courage are attributes that all judges in all ages and among all races have been supposed to possess in order to be just and righteous in their judgments. Disqualifications.—Disqualifications of applicants for membership in the Great Sanhedrin are not less interesting than qualifications. They are in the main mere negatives of affirmatives which have already been given, and would seem, therefore, to be superfluous. But they are strongly accentuated in Hebrew law, and are therefore repeated here. (1) A man was disqualified to act as judge who had not, or had never had, any regular trade, occupation, or profession by which he gained his livelihood. The reason for this disqualification was based upon a stringent maxim of the Rabbins: "He who neglects to teach his son a trade, is as though he taught him to steal!" A man who did not work and had never labored in the sweat of his brow for an honest livelihood, was not qualified, reasoned the Hebrew people, to give proper consideration or extend due sympathy to the cause of litigants whose differences arose out of the struggles of everyday life.
- 84. (2) In trials where the death penalty might be inflicted, an aged man, a person who had never had any children of his own, and a bastard were disqualified to act as judge. A person of advanced years was disqualified because according to the Rabbins old age is frequently marked by bad temper; and "because his years and infirmities were likely to render him harsh, perhaps obstinate and unyielding." On the other hand, youth was also a disqualification to sit in the Sanhedrin. According to the Rabbis, twenty-five years was the age which entitled a person to be called a Man;[92] but no one was eligible to a seat in the Sanhedrin until he had reached the age of forty years.[93] The ancient Hebrews regarded that period as the beginning of discretion and understanding. A person without children was not supposed to possess those tender paternal feelings "which should warm him on behalf of the son of Israel who was in peril of his life." The stain of birth and the degradation in character of a bastard were wholly inconsistent with the high ideals of the qualifications of a Hebrew judge. (3) Gamblers, dice players, bettors on pigeon matches, usurers, and slave dealers were disqualified to act as judges. The Hebrews regarded gambling, dice playing, betting on pigeon matches, and other such practices as forms of thievery; and thieves were not eligible to sit as judges in their courts. No man who was in the habit of lending money in an usurious manner could be a judge. It was immaterial whether the money was lent to a countryman or a stranger. Slave dealers were disqualified to act as judges because they were regarded as inhuman and unsympathetic. (4) No man was qualified to be a judge who had dealt in the fruits of the seventh year.
- 85. Such a person was deemed lacking in conscience and unfitted to perform judicial functions. (5) No man who was concerned or interested in a matter to be adjudicated was qualified to sit in judgment thereon. This is a universal disqualification of judges under all enlightened systems of justice. The weakness and selfishness of human nature are such that few men are qualified to judge impartially where their own interests are involved. (6) All relatives of the accused man, of whatever degree of consanguinity, were disqualified from sitting in judgment on his case. This is only a variation of the disqualification of interest. (7) No person who would be benefited, as heir, or otherwise, by the death or condemnation of an accused man, was qualified to be his judge. This, too, is a variation of the disqualification of interest. (8) The king could not be a member of the Sanhedrin. Royalty disqualified from holding the place of judge because of the high station of the king and because his exercising judicial functions might hamper the administration of justice. And, finally, in closing the enumeration of disqualifications, it may be added that an election to a seat obtained by fraud or any unfair means was null and void. No respect was shown for the piety or learning of such a judge; his judicial mantle was spat upon with scorn, and his fellow judges fled from him as from a plague or pest. Hebrew contempt for such a judge was expressed in the maxim: "The robe of the unfairly elected judge is to be respected not more than the blanket of an ass." Officers of the Great Sanhedrin.—Two presiding officers directed the proceedings of the Great Sanhedrin. One of these, styled prince
- 86. (nasi), was the chief and the president of the court. The other, known as the father of the Tribunal (ab-beth-din), was the vice- president. There has been much discussion among the historians as to the particular chamber from which the president was chosen. Some have contended that the presidency of the Sanhedrin belonged by right to the high priest. But the facts of history do not sustain this contention. Aaron was high priest at the time when Moses was president of the first Sanhedrin in the Wilderness; and, besides, the list of presidents preserved by the Talmud reveals the names of many who did not belong to the priesthood. Maimonides has made the following very apt observation on the subject: "Whoever surpassed his colleagues in wisdom was made by them chief of the Sanhedrin."[94] According to most Jewish writers, there were two scribes or secretaries of the Sanhedrin. But several others contend that there were three. Benny says: "Three scribes were present; one was seated on the right, one on the left, the third in the center of the hall. The first recorded the names of the judges who voted for the acquittal of the accused, and the arguments upon which the acquittal was grounded. The second noted the names of such as decided to condemn the prisoner and the reasons upon which the conviction was based. The third kept an account of both the preceding so as to be able at any time to supply omissions or check inaccuracies in the memoranda of his brother reporters."[95] In addition to these officers, there were still others who executed sentences and attended to all the police work of legal procedure. They were called shoterim.[96] There was no such officer as a public prosecutor or State's attorney known to the laws of the ancient Hebrews. The witnesses to the crime were the only prosecutors recognized by Hebrew criminal jurisprudence; and in capital cases they were the legal executioners as well.
- 87. There was also no such body as the modern Grand Jury known to ancient Hebrew criminal law. And no similar body of committee of the Sanhedrin performed the accusatory functions of the modern Grand Jury. The witnesses were the only accusers, and their testimony was both the indictment and the evidence. Until they testified, the man suspected was deemed not only innocent but unaccused. The profession of the law, in the modern sense of the term, was no part of the judicial system of the ancient Hebrews. There were no advocates as we know them. There were, indeed, men learned in the law—Pharisees and Sadducees—who knew all the law. There were doctors of the law: men whom Jesus confounded when a youth in the Temple at the age of twelve.[97] But there were no lawyers in the modern sense: professional characters who accept fees and prosecute cases. The judges and disciples performed all the duties of the modern attorney and counselor-at-law. The prophets were the sole orators of Hebrew life, but they were never allowed to appear as defendants of accused persons. Indeed, they themselves were at times compelled to play the role of defendants. Jeremiah is an illustrious example.[98] Both Keim[99] and Geikie[100] speak of a Baal Rib, a counsel appointed to see that everything possible was done to secure the rights of an accused person at a Hebrew criminal trial. But these statements are not in accord with standard works on ancient Hebrew jurisprudence. Indeed, Friedlieb emphatically denies that there was any such person as a Baal Rib or Dominus Litis among the ancient Hebrews.[101] It seems that in the closing years of Jewish nationality, specially retained advocates were known, for St. Luke tells us that the Jews employed Tertullus, a certain orator, to prosecute St. Paul.[102] But this was certainly an exceptional case. It is historically certain that in the early ages of the Jewish Commonwealth litigants pleaded their own causes. This we learn from the case of the two women who appeared before King Solomon, and laid before him their respective claims to a child.[103]
- 88. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com