![Microbes: An Unseen Majority Around Us | 21
sequencing for detecting candidate infectious agents
associated with human tissue. PLoS One 6, e19838.
Moran, M.A. (2009). Metatranscriptomics: eavesdrop-
ping on complex microbial communities. Microbe 4,
329–335.
Moreno-Paz, M., Gómez, M.J., Arcas, A., and Parro,
V. (2010). Environmental transcriptome analysis
reveals physiological differences between biofilm and
planktonic modes of life of the iron oxidizing bacteria
Leptospirillum spp. in their natural microbial commu-
nity. BMC Genomics 11, 404.
Motti, C. (2012). Environmental marine metabolomics:
from whole organism system biology to ecosystem
management. J. Mar. Sci. Res. Dev. 2, 3.
Moyer, C.L., and Morita, R.Y. (2007). Psychrophiles and
Psychrotrophs. In Encyclopedia Life Sciences (John
Wiley Sons Ltd, Chichester, UK). Available from:
http://www.els.net [doi: 10.1002/9780470015902.
a0000402.pub2].
Muth, T., Benndorf, D., Reichl, U., Rapp, E., and Martens,
L. (2012). Searching for a needle in a stack of needles:
challenges in metaproteomics data analysis. Mol. Bio-
syst. 9, 578–585.
Naeem, S., and Li, S.B. (1997). Biodiversity enhances
ecosystem reliability. Nature 390, 507–509.
Nagalakshmi, U., Wang, Z., Waern, K., Shou, C., Raha, D.,
Gerstein, M., and Snyder, M. (2008). The transcrip-
tional landscape of the yeast genome defined by RNA
sequencing. Science 320, 1344–1349.
Newman, D.J., and Cragg, G.M. (2007). Natural products
as sources of new drugs over the last 25 years. J. Nat.
Prod. 70, 461–477.
Ohkuma, M. (2003). Termite symbiotic systems: efficient
bio-recycling of lignocellulose. Appl. Microbiol. Bio-
technol. 61, 1–9.
Ortmann, A.C., and Suttle, C.A. (2009). Determination
of virus abundance by epifluorescence microscopy.
Methods Mol. Biol. 501, 87–95.
Oshlack, A., and Wakefield, M.J. (2009). Transcript length
bias in RNAseq data confounds systems biology. Biol.
Direct. 4, 14.
Pace, N.R. (1997). A molecular view of microbial diversity
and the biosphere. Science 276, 734–740.
Pan, S.Y., Pan, S., Yu, Z.L., Ma, D.L., Chen, S.B., Fong, W.F.,
Han, Y.F., and Ko. K.M.. (2010). New perspectives
on innovative drug discovery: an overview. J. Pharm.
Pharm. Sci. 13, 450–71.
Parro, V., Moreno-Paz, M., and González-Toril, E. (2007).
Analysis of environmental transcriptomes by DNA
microarrays. Environ. Microbiol. 9, 453–64.
Passalacqua, K.D., Varadarajan, A., Ondov, B.D., Okou,
D.T., Zwick, M.E., and Bergman, N.H. (2009) Struc-
ture and complexity of a bacterial transcriptome. J.
Bacteriol. 191, 3203–3211.
Petrić, I., Bru, D., Udiković-Kolić, N., Hršak, D., Philippot,
L., and Martin-Laurent F. (2011). Evidence for shifts
in the structure and abundance of the microbial com-
munity in a long-term PCB-contaminated soil under
bioremediation. J. Hazard Mater. 195, 254–60.
Petti, C.A. (2007). Detection and identification of micro-
organisms by gene amplification and sequencing. Clin.
Infect. Dis. 44, 1108–1114.
Pinto, A.C., Melo-Barbosa, H.P., Miyoshi, A., Silva, A., and
Azevedo, V. (2011). Application of RNA-seq to reveal
the transcript profile in bacteria. Genet. Mol. Res. 10,
1707–18.
Poretsky, R.S., Hewson, I., Sun, S., Allen, A.E., Zehr, J.P.,
and Moran, M.A. (2009). Comparative day/night
metatranscriptomic analysis of microbial communities
in the North Pacific Subtropical Gyre. Environ. Micro-
biol. 11, 1358–1375.
Porter, K.G., and Feig, Y.S. (1980). The use of DAPI for
identifying and counting aquatic microflora. Limnol.
Oceanogr. 25, 943–948.
Prakash, T., and Taylor, T.D. (2012). Functional
assignment of metagenomic data: challenges and
applications. Brief. Bioinf. 13, 711–27.
Psenner, R. (1990). From image analysis to chemical anal-
ysis of bacteria: a long-term study? Limnol. Oceanogr.
35, 234–237.
Ram, R.J., Verberkmoes, N.C., Thelen, M.P., Tyson,
G.W., Baker, B.J., Blake II, R.C., Shah, M., Hettich,
R.L., and Banfield, J.F. (2005). Community prot-
eomics of a natural microbial biofilm. Science 308,
1915–1920.
Rappe, M.S., and Giovannoni, S.J. (2003). The uncul-
tured microbial majority. Annu. Rev. Microbiol. 57,
369–394.
Rasmussen, S., Nielsen, H.B., and Jarmer, H. (2009). The
transcriptionally active regions in the genome of Bacil-
lus subtilis. Mol. Microbiol. 73, 1043–1057.
Raz, T., Kapranov, P., Lipson, D., Letovsky, S., Milos, P.M.,
and Thompson, J.F. (2011). Protocol dependence of
sequencing-based gene expression measurements.
PLoS One 6, e19287.
Roh, S.W., Abell, G.C., Kim, K.H., Nam, Y.D., and Bae,
J.W. (2010). Comparing microarrays and next-gener-
ation sequencing technologies for microbial ecology
research. Trends Biotechnol. 28, 291–299.
Roselló-Mora, R., and Amann, R. (2001). The species
concept for prokaryotes. FEMS Microbiol. Rev. 25,
39–67.
Rothberg, J.M., and Leamon, J.H. (2008). The develop-
ment and impact of 454 sequencing. Nat. Biotechnol.
26, 1117–1124.
Rothschild, L.J., and Mancinelli, R.L. (2001). Life in
extreme environments. Nature 409, 1092–1101.
Sanger, F., Nicklen, S., and Coulson, A.R. (1977). DNA
sequencing with chain-terminating inhibitors, Proc.
Natl. Acad. Sci. U.S.A. 74, 5463–5467.
Schleifer, K.H. (2009). Classification of Bacteria and
Archaea: past, present and future. Syst. Appl. Micro-
biol. 32, 533–542.
Schmid, A., Dordick, J.S., Hauer, B., Kiener, A., Wubbolts,
M., and Witholt, B. (2001). Industrial biocatalysis
today and tomorrow. Nature 409, 258–268.
Schmidt, T.M., DeLong, E.F., and Pace, N.R. (1991).
Analysis of a marine picoplankton community by 16S
rRNA gene cloning and sequencing. J. Bacteriol. 173,
4371–4378.
Schneider, M.V., and Orchard, S. (2011). Omics technolo-
gies, data and bioinformatics principles. Methods Mol.
Biol. 719, 3–30.](https://image.slidesharecdn.com/2651750-250527072326-128e993d/85/Bioinformatics-And-Data-Analysis-In-Microbiology-Ozlem-Tastan-Bishop-36-320.jpg)
![Graham et al.
26 |
In this chapter we provide an overview of
sequencing technologies as well as the down-
streamprocessingstepsthatcanbeusedtoconvert
the raw reads into larger contigs suitable for anno-
tation of genes and other genomic features (Fig.
2.1). Strategies for specific sequencing objectives
are presented, which include suggestions for
sequencing technology, library types, assembly
methods and software. Given the rapid rate at
which new sequencing technologies and bioinfor-
matics tools are being produced, such strategies
need to be regularly revised, but these guidelines
should serve as a useful starting point. Lastly, the
chapter is intended to introduce the method-
ologies associated with genome sequencing and
assembly; it not intended to be comprehensive
with respect to existing bioinformatics tools.
Sequencing technologies
First generation sequencing
approaches
The first widely adopted approaches to DNA
sequencing were developed in the mid-seventies.
The pioneering method invented by Frederick
Sanger and Alan Coulsen (Sanger and Coulson,
1975), commonly referred to as the Sanger method
or chain-termination sequencing, is based on the
incorporation of chain-terminating 2′,3′-dide-
oxyribonucleotide 5′-triphosphates (also called
dideoxynucleotides; ddNTPs) during replication of
a single-stranded DNA template fragment. In this
approach, the DNA fragment to be sequenced is
separated into four different reactions containing
radiolabelled DNA primer, DNA polymerase,
and one of the four dideoxynucleotides (con-
taining adenosine (ddATP), cytosine (ddCTP),
guanosine (ddGTP), or thymidine (ddTTP),
respectively). A dideoxynucleotide terminates
the chain thereby blocking extension because
other nucleotides cannot bind to it owing to
its 3′-end modification. Thus, the polymerase
extends the labelled DNA primer until a dide-
oxynucleotide is randomly incorporated, at which
point the sequencing reaction is terminated.
These fragments are then denatured, separated by
electrophoresis and visualized using radiography.
The random incorporation of chain-terminating
ddNTPs guarantees that every fragment length
(correspondingtoeverypossiblesinglenucleotide
addition) will be represented as a distinct band in
the electropherogram, thus providing a mecha-
nism to deduce the genomic sequence. A different
approach, developed around the same time by
Alan Maxam and Walter Gilbert (Maxam and Gil-
bert, 1977) employs radiolabelling at the 5′ end of
the DNA fragment to be sequenced, followed by
chemical treatments to cleave the DNA fragment
at specific residues or residue pairs. The fragments
are separated by size using electrophoresis, and
then detected by autoradiography. Analysis of
.
.
.
.
.
.
.
.
.
.
Process
Time
Frame
Microbial
WGS
Project
Flow
Figure 2.1 #5 /%
* ]# #5
#5 *
^ # study design
I * 7
# /%
/%
/% : ) 7
# D$ 7 %
_ ! /%7
7 Sample collection
!
!7 ) ` !
7# . I
#
! !
/% '% D$
/% culture
7 % 5
% % ! NGS
library construction sequencing are described
] :00 7data analysis7
% j 7 %
) /% % %
:0](https://image.slidesharecdn.com/2651750-250527072326-128e993d/85/Bioinformatics-And-Data-Analysis-In-Microbiology-Ozlem-Tastan-Bishop-41-320.jpg)
![Graham et al.
28 |
the growing strand. This can be accomplished by
supplying substrates (nucleotides or short oligo-
nucleotides) that are modified with an identifying
tag (such as a fluorophore) and also reversibly
blocked or else by providing only a single kind of
substrate (e.g. dATP) at a time. Nucleotides and
reagents are cyclically and individually washed
across the immobilized templates, whereupon the
specific incorporation of nucleotide(s) is detected
by high-resolution digital imaging. Thereafter,
reagents are washed away and a new reagent cycle
is initiated. Thus, most NGS is currently achieved
on spatially distinct, immobilized DNA template
libraries via cycles of nucleotide incorporation/
washing/imaging. In contrast, single-molecule
sequencing (SMS) requires no such nucleotide/
washing/imaging cycles; instead SMS directly
monitors and detects nucleotide addition on a
single DNA fragment strand. Other proprietary
features of each NGS technology are related to the
specific automated fluidic and optic technologies
applied to capture each nucleotide series for each
template library. Lastly, it is worth mentioning
that regeneration of raw image data is less expen-
sive than storage of raw data owing to the sheer
size of NGS image data files produced; hence the
collected raw data files are generally deleted once
processed.
All current commercial NGS technologies
require DNA template preparation (NGS library
construction), immobilization of said libraries,
followed by massively parallel sequencing. The
overall NGS procedure is summarized in Fig.
2.2. Proprietary nuances rest in how the stages
are achieved. In general terms, bacterial genome
sequencing is accomplished by highly redundant/
deep sequencing of random fragments making up
the genome (Fig. 2.3). Random genomic library
sequencing is sometimes referred to as ‘shotgun
sequencing’, and is often achieved with ‘single-end’
sequencing of the library fragments (although
not always – see below). In single-end sequenc-
ing, the library fragment is sequenced exclusively
from one direction. Random fragmentation of
genomic DNA is achieved using mechanical
shearing, sonication or enzymatic processes. The
random DNA fragment ends are then made ame-
nable to immobilization and/or sequencing via
end polishing and incorporation of universal but
platform-specific sequences (called ‘adaptors’).
Following immobilization, each adaptor-contain-
ing template library most often requires clonal
amplification to achieve sufficient quantities prior
to massively parallel sequencing and detection.
Amplification can be performed on tethered
libraries or in solution, but the efficiency of any
NGS platform is tied to the relative assurance that
each amplified library clone contains multiple
copies of only a single and identical DNA frag-
ment. The sole exception is the single-molecule
sequencing approach commercialized as PacBio
RS by Pacific Biosciences, which as the name sug-
gests operates on a single DNA fragment.
Modification of the standard ‘shotgun’ tem-
plate library procedure can be applied to generate
mate-pair libraries, comprised of originally dis-
continuous sequence fragments from the genome
separated by intervening nucleotides that are later
enzymatically linked together in order to undergo
sequencing. As such, mate-pair sequencing
™ TEMPLATE QC PREPARATION
(randomly shear; end repair; size select)
™ NGS LIBRARY CONSTRUCTION
(append sequencing adaptors + multiplex identifiers)
™ CLONAL AMPLIFICATION of Libraries
(layout library on sequencing slide/wells)
™ MASSIVELY PARALLEL SEQUENCING
(determine order identity of bases [at either end] of fragment)
™ RAW DATA ANALYSIS
(image processing + base calling, with associated qualities)
Figure 2.2 D) 5 /%
!!#I 7 ) 5 *
% . % I
template QC and preparation7D
./%
# I NGS library construction7
D/% !
/% = I clonal
7
%
% %/%
/% D =
/%$$/% I massively parallel
sequencing !%
/% z raw data analysis7
/% %
/% D$ #
/% % %](https://image.slidesharecdn.com/2651750-250527072326-128e993d/85/Bioinformatics-And-Data-Analysis-In-Microbiology-Ozlem-Tastan-Bishop-43-320.jpg)
![Genome Sequencing and Assembly | 29
provides a cost-efficient means of co-sequencing
discontinuous fragments (mates) from across
the genome. Knowing that the mated reads must
face each other (in most cases) across an average
known gap size, even without determining the
intervening sequence within the gap the mated
reads can be applied to generate a ‘genomic scaf-
fold’ that is useful for orienting gapped contigs
(Fig. 2.3). Paired-end sequencing refers to the
sequencing of both ends of DNA library frag-
ments and is thus performed when mate-pair
libraries are used. Paired-end sequencing can also
be applied to fragments prepared through stand-
ard, non-mate-pair (i.e. shotgun) template library
procedures. In this case, the paired reads are from
each end of the same NGS library fragment, and
are thus separated on the original genome by
only the small NGS fragment insert size. In the
literature the term ‘paired-end’ is sometimes used
to refer exclusively to this latter scenario; some-
times for long-insert, mate-pair libraries (Roche
pyrosequencing). Owing to differences in library
Paired-end
sequencing
Genomic DNA
Generate random
genomic fragments
Genomic scaffold
Mate-pair reads
Contigs
Single-end
sequencing
“draft” genome
“finished” genome
Gap-filling
Sequence polishing
De novo assembly
Mate pair library
protocol
NGS library construction
Figure 2.3 z 7 7 =
=%% D$ /% D
! D$ 7
=D{ | % .D$ $/% % single-
endpaired-end/% # # {mate-pair|
# 5 /% # 7 5 %%
# ! % . # /% ) . # 5
% %| % 5 5
:;; de novo assembled
{ |% { |
^7 % 7% % * 7# in
silico . Y) ! %
# ! (' = # % 5
D # . `% ! ] . /%
% ' ;7:;$% % /% %
_/% 7# !=!
{ |' %%= %
) 7%](https://image.slidesharecdn.com/2651750-250527072326-128e993d/85/Bioinformatics-And-Data-Analysis-In-Microbiology-Ozlem-Tastan-Bishop-44-320.jpg)
![This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.
Title: Kaksintaistelu
Author: Anton Pavlovich Chekhov
Translator: Emil Mannstén
Release date: February 17, 2013 [eBook #42116]
Language: Finnish
Credits: Produced by Jukka Aakula
*** START OF THE PROJECT GUTENBERG EBOOK KAKSINTAISTELU
***](https://image.slidesharecdn.com/2651750-250527072326-128e993d/85/Bioinformatics-And-Data-Analysis-In-Microbiology-Ozlem-Tastan-Bishop-60-320.jpg)
![Produced by Jukka Aakula
KAKSINTAISTELU
Kirj.
Anton Tshehov
Venäjänkielestä [Duelj] suomentanut Emil Mannstén
Arvi A. Karisto Oy, Hämeenlinna, 1921.](https://image.slidesharecdn.com/2651750-250527072326-128e993d/85/Bioinformatics-And-Data-Analysis-In-Microbiology-Ozlem-Tastan-Bishop-61-320.jpg)
Bioinformatics And Data Analysis In Microbiology Ozlem Tastan Bishop
- 1. Bioinformatics And Data Analysis In Microbiology Ozlem Tastan Bishop download https://ebookbell.com/product/bioinformatics-and-data-analysis- in-microbiology-ozlem-tastan-bishop-5303500 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. Advances In Data Analysis Theory And Applications To Reliability And Inference Data Mining Bioinformatics Lifetime Data And Neural Networks 1st Edition Ramn Alvarezesteban https://ebookbell.com/product/advances-in-data-analysis-theory-and- applications-to-reliability-and-inference-data-mining-bioinformatics- lifetime-data-and-neural-networks-1st-edition-ramn- alvarezesteban-4259894 Applying Big Data Analytics In Bioinformatics And Medicine 1st Edition Paraskevi Papadopoulou Editor https://ebookbell.com/product/applying-big-data-analytics-in- bioinformatics-and-medicine-1st-edition-paraskevi-papadopoulou- editor-42166254 Big Data Analytics In Chemoinformatics And Bioinformatics With Applications To Computeraided Drug Design Cancer Biology Emerging Pathogens And Computational Toxicology Subhash C Basak https://ebookbell.com/product/big-data-analytics-in-chemoinformatics- and-bioinformatics-with-applications-to-computeraided-drug-design- cancer-biology-emerging-pathogens-and-computational-toxicology- subhash-c-basak-47921176 Advances In Bioinformatics And Big Data Analytics Sujata Dash https://ebookbell.com/product/advances-in-bioinformatics-and-big-data- analytics-sujata-dash-50738320
- 3. Computational Intelligence And Big Data Analytics Applications In Bioinformatics 1st Ed Ch Satyanarayana https://ebookbell.com/product/computational-intelligence-and-big-data- analytics-applications-in-bioinformatics-1st-ed-ch- satyanarayana-7328082 Bioinformatics And Biomarker Discovery Omic Data Analysis For Personalized Medicine 1st Edition Francisco Azuaje https://ebookbell.com/product/bioinformatics-and-biomarker-discovery- omic-data-analysis-for-personalized-medicine-1st-edition-francisco- azuaje-1339810 Big Data Analysis For Bioinformatics And Biomedical Discoveries Ye https://ebookbell.com/product/big-data-analysis-for-bioinformatics- and-biomedical-discoveries-ye-5308452 Big Data Analysis For Bioinformatics And Biomedical Discoveries Shui Qing Ye Editor https://ebookbell.com/product/big-data-analysis-for-bioinformatics- and-biomedical-discoveries-shui-qing-ye-editor-9970720 Bioinformatics A Practical Guide To Next Generation Sequencing Data Analysis Chapman Hallcrc Computational Biology Series 1st Edition Hamid D Ismail https://ebookbell.com/product/bioinformatics-a-practical-guide-to- next-generation-sequencing-data-analysis-chapman-hallcrc- computational-biology-series-1st-edition-hamid-d-ismail-51574846
- 5. Bioinformatics and Data Analysis in Microbiology Edited by Caister Academic Press
- 6. Caister Academic Press Bioinformatics and Data Analysis in Microbiology Edited by ! # $%
- 7. Copyright © 2014 Caister Academic Press Norfolk, UK www.caister.com British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-908230-39-3 (hardback) ISBN: 978-1-908230-73-7 (ebook) Description or mention of instrumentation, software, or other products in this book does not imply endorsement by the author or publisher. The author and publisher do not assume responsibility for the validity of any products or procedures mentioned or described in this book or for the consequences of their use. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher. No claim to original U.S. Government works. Cover design adapted from Figures 7.1 and 9.2.
- 8. Contents ' % ! ( ) 1 * % + ,!!# - Meesbah Jiwaji, Gwynneth F. Matcher and Rosemary A. Dorrington 2 (. $/% 03 Morag Graham, Gary Van Domselaar and Paul Stothard 3 5#6 ! $/% 7 (% 5# 3- 4 (. 9- Gary Van Domselaar, Morag Graham and Paul Stothard 3 ' ! ( $ % /% : --; Jade Hotchkiss and Nicola J. Mulder 6 '% (= +'% :% % -; Angel Valverde, Pieter De Maayer and Don A. Cowan 7 -@; ! # # # $% 9 B% ! -@D -F; ' ( ) * )+ 9 P 0GH Oleg Paliy, Vijay Shankar and Marketa Sagova-Mareckova
- 9. Contents iv | -G 00F /%01 2 3 /1#4 1 56 6 * 7 1 68 5*11 0 I ) 03
- 10. Contributors Lilija V. Avdeeva Department of Antibiotics Institute of Microbiology and Virology NAS of Ukraine Kiev Ukraine avdeeva@imv.kiev.ua Sofia Barreira School of Maths, Applied Maths and Statistics College of Science NUI Galway Ireland s.depereirabarreira1@nuigalway.ie Oliver K.I. Bezuidt Department of Biochemistry, Bioinformatics and Computational Biology Unit University of Pretoria Pretoria South Africa bezuidt@gmail.com Rainer Borriss ABiTEP CmbH Glienicker Weg 185 Berlin Germany rborriss@abitep.de Wai Y. Chan Department of Microbiology and Plant Pathology University of Pretoria Pretoria South Africa Annie.Chan@fabi.up.ac.za Simone Coughlan School of Maths, Applied Maths and Statistics College of Science NUI Galway Ireland s.coughlan9@nuigalway.ie Don A. Cowan Centre for Microbial Ecology and Genomics Department of Genetics University of Pretoria Pretoria South Africa don.cowan@up.ac.za Pieter De Maayer Centre for Microbial Ecology and Genomics Department of Genetics University of Pretoria Pretoria South Africa pieter.demaayer@fabi.up.ac.za
- 11. Contributors vi | Rosemary A. Dorrington Department of Biochemistry and Microbiology Rhodes University Grahamstown South Africa r.dorrington@ru.ac.za Tim Downing School of Maths, Applied Maths and Statistics College of Science NUI Galway Ireland tim.downing@nuigalway.ie Anthony Fodor Department of Bioinformatics and Genomics UNC Charlotte Charlotte, NC USA afodor@uncc.edu Johannes B. Goll The EMMES Corporation N Washington St Rockville, MD USA johannes.goll@gmail.com Morag Graham National Microbiology Laboratory Public Health Agency of Canada Winnipeg, MB Canada morag.graham@phac-aspc.gc.ca Jade Hotchkiss Institute of Infectious Disease and Molecular Medicine Faculty of Health Sciences University of Cape Town Cape Town South Africa giant.plankton@gmail.com Meesbah Jiwaji Department of Biochemistry and Microbiology Rhodes University Grahamstown South Africa m.jiwaji@ru.ac.za Konstantinos Krampis The J. Craig Venter Institute Rockville MD USA agbiotec@gmail.com Svitlana V. Lapa Department of Antibiotics Institute of Microbiology and Virology NAS of Ukraine Kiev Ukraine slapa@ukr.net Jonathan McCafferty Department of Bioinformatics and Genomics UNC Charlotte Charlotte, NC USA jmccaff2@uncc.edu Gwynneth F. Matcher Department of Biochemistry and Microbiology Rhodes University Grahamstown South Africa g.matcher@ru.ac.za Nicola J. Mulder Institute of Infectious Disease and Molecular Medicine Faculty of Health Sciences University of Cape Town Cape Town South Africa nicola.mulder@uct.ac.za
- 12. Contributors | vii Karen E. Nelson The J. Craig Venter Institute Rockville, MD USA kenelson@jcvi.org Oleg Paliy Boonshoft School of Medicine Wright State University Dayton, OH USA oleg.paliy@wright.edu Oleg N. Reva Department of Biochemistry, Bioinformatics and Computational Biology Unit University of Pretoria Pretoria South Africa oleg.reva@up.ac.za Larisa A. Safronova Department of Antibiotics Institute of Microbiology and Virology NAS of Ukraine Kiev Ukraine safronova_larisa@ukr.net Marketa Sagova-Mareckova Crop Research Institute Prague Czech Republic sagova@vurv.cz Cathal Seoighe School of Maths, Applied Maths and Statistics College of Science NUI Galway Ireland cathal.seoighe@nuigalway.ie Vijay Shankar Boonshoft School of Medicine Wright State University Dayton, OH USA shankar.5@wright.edu Paul Stothard Department of Agricultural, Food and Nutritional Science University of Alberta Edmonton, AB Canada stothard@ualberta.ca Sebastian Szpakowski The J. Craig Venter Institute Rockville, MD USA shpakoo@gmail.com Özlem Taştan Bishop Research Unit in Bioinformatics (RUBi) Department of Biochemistry and Microbiology Rhodes University Grahamstown South Africa ozlem.tastanbishop@gmail.com Angel Valverde Centre for Microbial Ecology and Genomics Department of Genetics University of Pretoria Pretoria South Africa angel.valverde@up.ac.za Gary Van Domselaar National Microbiology Laboratory Public Health Agency of Canada Winnipeg, MB Canada gary.vandomselaar@phac-aspc.gc.ca
- 13. Current books of interest Microarrays: Current Technology, Innovations and Applications 2014 Metagenomics of the Microbial Nitrogen Cycle: Theory, Methods and Applications 2014 Proteomics: Targeted Technology, Innovations and Applications 2014 Biofuels: From Microbes to Molecules 2014 Human Pathogenic Fungi: Molecular Biology and Pathogenic Mechanisms 2014 Applied RNAi: From Fundamental Research to Therapeutic Applications 2014 Halophiles: Genetics and Genomes 2014 Phage Therapy: Current Research and Applications 2014 The Cell Biology of Cyanobacteria 2014 Pathogenic Escherichia coli: Molecular and Cellular Microbiology 2014 Campylobacter Ecology and Evolution 2014 Burkholderia: From Genomes to Function 2014 Myxobacteria: Genomics, Cellular and Molecular Biology 2014 Next-generation Sequencing: Current Technologies and Applications 2014 Omics in Soil Science 2014 Applications of Molecular Microbiological Methods 2014 Mollicutes: Molecular Biology and Pathogenesis 2014 Genome Analysis: Current Procedures and Applications 2014 Bacterial Toxins: Genetics, Cellular Biology and Practical Applications 2013 Bacterial Membranes: Structural and Molecular Biology 2014 Cold-Adapted Microorganisms 2013 Fusarium: Genomics, Molecular and Cellular Biology 2013 Prions: Current Progress in Advanced Research 2013 RNA Editing: Current Research and Future Trends 2013 Real-Time PCR: Advanced Technologies and Applications 2013 Microbial YZ%) Pumps: Current Research 2013 Cytomegaloviruses: From Molecular Pathogenesis to Intervention 2013 Oral Microbial Ecology: Current Research and New Perspectives 2013 Bionanotechnology: Biological Self-assembly and its Applications 2013 Real-Time PCR in Food Science: Current Technology and Applications 2013 Bacterial Gene Regulation and Transcriptional Networks 2013 Bioremediation of Mercury: Current Research and Industrial Applications 2013 Neurospora: Genomics and Molecular Biology 2013 Rhabdoviruses 2012 Full details at www.caister.com
- 14. Preface We are at the door of an exciting future for micro- biology. The rapid advancement of sequencing techniques, coupled with the new methodologies of bioinformatics to handle large scale data analy- sis, are providing exciting opportunities for us to understand microbial communities from a variety of environments beyond previous imagination. Data analysis is extremely important for a deeper knowledge of microbes and their habitats, and for many applications of microbiology ranging from understanding the basis of diseases or host pathogen interactions so as to design drugs and develop vaccines, to many other biotechnology applications, including barcoding, microbial bio- remediation and bio-fuel production. This book aims to present up-to-date and detailed information on various aspects of bio- informatics data analysis with applications to microbiology. It describes a number of different useful bioinformatics tools, highlights the links to some wet-lab techniques, explains different approaches to tackle a problem, talks about cur- rent challenges and limitations, demonstrates the applications, and discusses future trends. A brief summary of each chapter is as follows: Chapter 1 provides a review of microbes and their importance in the context of ecosystems, and an overview of the methods applied to study microbes within ecosystems. It aims to give an introduction to newcomers to the fields of micro- biology and bioinformatics. Chapter 2 reviews sequencing technologies and discusses the challenges of assembly and ways of tackling the problems.Chapter3highlightsthescopeofmicro- bial variations and explores the importance of accurate genome assembly for structural variation and single nucleotide polymorphism analysis. Chapter 4 deals with genome annotation, and reviews computational methods that have been developed for annotation of bacterial and archaeal genomes. Chapter 5 explores the methods for comparative analysis of microbial genomes, and focuses on the analysis of Mycobacteria as a case study. Microbial community profiling is the topic of Chapter 6, in which fingerprinting techniques and construction of phylogenetic trees are interro- gated. Metagenomics, one of the fastest advancing fields of microbiology, is discussed in Chapter 7 with examples of the Human Microbiome Project and a Baltic Sea study. Chapter 8 further inter- rogates human microbiome analysis and presents the pros and cons of using 16S rRNA based sequencing studies. Chapter 9 takes us to another interesting topic, phylogenetic microarrays, in which 16S rRNA sequences are used to determine the composition of microbial communities, yet from the microarray aspect. The last chapter is about genetic barcoding with its applications to microbiology and biotechnology. Overall, this is an essential book for researchers, lecturers and students involved in microbiology and/or different aspects of bioinformatics including next- generation sequencing, sequence data analysis, comparative genome analysis, metagenomics and more. The book has been peer reviewed. Each chap- ter of the book has been reviewed by at least one reviewer and by the editor. The reviewing process was undertaken by other authors contributing to the book, and I believe the constructive com- ments of the reviewers significantly improved the quality of the book. I thank the authors of
- 15. Preface x | the book chapters for their participation in this process. My special thanks go to Tim Downing who was always available for the reviewing pro- cess, and always responded exceptionally fast. I am especially grateful to Gary Van Domselaar, Morag Graham and Paul Stothard who agreed to write an extra chapter in a very short time, when another author dropped out. I also would like to thank Fourie Joubert for his initial contribution in discussing the chapter topics and suggesting some potential authors. My deepest thanks to the Sabanci University, Istanbul-Turkey, for hosting me and providing a friendly working environment during my sabbatical which made it possible to finalize the book. Last but not least, I would like to thank my husband, Nigel T. Bishop, for his con- tinuous support in my first time as an editor. Özlem Taştan Bishop
- 16. 1 Understanding the Unseen Majority Around Us: An Overview of '!' ' Meesbah Jiwaji, Gwynneth F. Matcher and Rosemary A. Dorrington Abstract Of all the living organisms on the planet, microor- ganisms are the most numerically abundant and diverse in nature. Despite their ubiquity, research- ers have only begun to understand the diversity profiles, metabolic functioning and potential economic value of these organisms. Classical investigation of microorganisms involves the culture and study of selected microbes in the laboratory setting. While this approach has yielded much information, there are two major drawbacks. Firstly, most microbes present in the environment are unculturable using currently available media/methodologies and, secondly, they focus on one, often attenuated, isolate/species and/or a set of genes at a time. To overcome these problems, researchers have focussed on the development of new technolo- gies that yield large, reliable and robust datasets in fields that include genomics, transcriptomics and proteomics. Importantly, the development of high-throughput sequencing technologies has dramatically advanced the analysis of microbial species diversity and their functioning within ecosystems. The large volumes of information- rich data require intelligent, and often repetitive, computational analysis, stressing the need for development of suitable bioinformatics analy- sis tools. This chapter provides an overview of microbes and the importance of why we need to understand them, as well as the methods applied to studying microbiota within ecosystems. Introduction The field of microbiology is undergoing a renais- sance. The discipline has changed dramatically since Antonie van Leeuwenhoek, the man who invented the microscope, watched bacteria that he recovered from his own teeth through his homemade microscope (van Leeuwenhoek, 1677). Traditional microbiology has focussed on the study of individual organisms, but now, while microorganisms and their processes still need to be understood at the level of the indi- vidual, increasingly, the aim is to understand the cumulative influence of microorganisms on the functioning of biological systems. As a conse- quence, microbiology has progressed from the study of individual genomes to that of microbial populationsinecosystems.Scientificobservations at the subcellular level are now being interpreted at rapidly increasing levels of complexity in order to gain a better understanding of the complex metabolic pathways within cells as well as their interactions in the context of the ecosystem. In order to gain understanding of system organiza- tion on such a large scale, new methodologies are constantly being developed, most of which generate extremely large sets of raw data which need to be curated and analysed using the tools of bioinformatics. What is a microorganism? The term ‘microorganism’ is used to describe any small organism, typically an entity that has a mass of less than 10–5 g and a length of less than 500μm, with a largest dimension of 100–150μm (Hughes Martiny et al., 2006; Karl, 2007). It is important to
- 17. Jiwaji et al. 2 | note that currently organisms are assigned to the ‘micro’ class, based only on their size. This ignores any differences in their evolutionary histories and metabolic capabilities. Given time, and with the increasing awareness of the diversity of microor- ganisms, this method of classification may need to be revisited. The classification of organisms involves placing them within collections/groups of related species. Traditionally, microorganisms were classified within the Domains Prokarya or Eukarya based on their morphology, the environment they were isolated from, the means by which they generated energy, their nutrient requirements and their mode of replication (Woese et al., 1990; Karl, 2007; Schleifer, 2009). Since the advent of molec- ular biology and the collection of sequence data, microorganisms are being classified into three Domains; the Bacteria, Archaea and Eukarya (Fig. 1.1)(Woese et al.,1990).BacteriaandArchaeaare grouped together as prokaryotes based on several cellular characteristics. These include the absence of a membrane surrounding the genome, the lack of introns in the encoded genes, the absence of intracellular organelles (e.g. chloroplasts, mito- chondria, etc.) and ribosomal subunits of 30S and 50S (eukaryotic subunits are larger at 60S and 40S). While Archaea and Bacteria do share several similarities, significant differences do exist. For example, bacterial cell walls contain peptidogly- can whilst archaeal cell walls do not and bacterial cell membranes contain ester links whilst those of Archaea have ether links. While Archaea and Bac- teria are both classified as prokaryotes, Archaea do in fact share a number of morphological char- acteristics and DNA sequence similarities with that of Eukaryotes prompting the speculation that Archaea and Eukarya diverged from bacteria before they diverged from one another (Fig. 1.1) (Zillig, 1991; Schleifer, 2009). Microbes in the world around us – the unseen majority Microorganisms are very old; they are thought to have inhabited the Earth for more than 3.5 billion years as evidenced by their presence in fossils, for example the Burgess Shale in Canada (Schopf, 2001). It is interesting that through their evolu- tionary history, microorganisms have remained simple and small and that their classification is often expressed in terms of their physiology and metabolism rather than their morphologies. Microorganisms are integral components of all ecosystems and there is no environment stud- ied to date in which microorganisms have not been found. From the extreme environments where temperatures exceed 100°C or fall below ! # # $ Figure 1.1 != 7 Y%. ! !!% !
- 18. Microbes: An Unseen Majority Around Us | 3 0°C, pHs that range from below 2 to over 10, or pressures over 100MPa, to every conceivable environment in between (Ulukanli and Digrak, 2002; Baker-Austin and Dopson, 2007; Moyer and Morita, 2007; Fang et al., 2010; Synowiecki, 2010). In the polar regions to the deserts, from the shallow to the deep oceans, from exotic envi- ronments to one’s own backyard, there exists an unexpected abundance and diversity of microor- ganisms (Pace, 1997). In terms of abundance, for example, one gram of soil may harbour up to 10 billion microorganisms potentially representing thousands of different species (Roselló-Mora and Amann, 2001). It is now well recognized that microorganisms are both taxonomically diverse and metabolically complex. In fact, microorgan- isms dominate their environments in terms of biomass, diversity and metabolic activity and are thus truly the ‘unseen majority’ (DeLong and Karl, 2005). Of particular interest is the role of the microor- ganism in ecosystem biology. Ecosystems consist of interconnected systems whose components interact over a broad range of physical and bio- logical states (DeLong, 2007). While several physico-chemical parameters may be similar between different ecosystems, the microbial com- munities within these ecosystems differ. Thus, while microorganisms are ubiquitous, their distri- bution is not uniform and their diversity profiles differ significantly between ecological niches. A strong body of evidence shows that the individual environment selects, and is partly responsible for, the spatial variation in the diversity and popula- tion structure of microorganisms (Hughes Martiny et al., 2006). In addition to the effect of a wide range of physico-chemical conditions on microbial diversity profiles, factors such as inter- species competition and predation also play a role in effecting relative abundances and distributions of microbial species (DeLong, 2007; Hibbing et al., 2010). Microorganisms form a considerable propor- tion of the total biomass in all ecosystems and so are supremely important in the functioning of global ecosystem processes. For example, in marine ecosystems, it is estimated that the microbes are responsible for up to 98% of the primary productivity and the mediation of all biogeochemical processes (Sogin et al., 2006). Microorganisms harvest and convert solar energy, catalyse important biogeochemical transforma- tionsoffreenutrientsandtraceelementstosustain ecosystems, they form crucial links in the carbon and nitrogen cycles, and they both produce and consume greenhouse gases including carbon dioxide (CO2 ), nitrous oxide (N2 O) and methane (CH4 ) (McGrady-Steed et al., 1997; Naeem and Li, 1997; van der Heijden et al., 1998; Cavigelli and Robertson, 2000; Horz et al., 2004; Bell et al., 2005). In addition, cellular microorganisms are pivotal in driving the sulfur and phosphorous cycles, the production of secondary metabolites including vitamins and co-factors and bioac- tive compounds such as antibiotics that confer selective advantages in the highly competitive environmental niches (Fagerbakke et al., 1996; Demain, 2007). In the past, the majority of microbial genetic datageneratedhasbeendefinedbytheresearcher’s individual areas of interest and has often focused on individual species of microorganisms. For example, the study of Acidithiobacillus ferrooxidans and its application to bioremediation (Umrania, 2006). More recently, there has been a shift in the approach to studies and there has been increasing interest in the study of environmental ecosystems as a whole. As examples, Desai et al. (2010), Xu et al. (2010) and Petrić et al. (2011) examined the microbial communities in xenobiotic-contami- nated soils during bioremediation. Microorganisms and climate change The study of climate change includes green- house-gas-induced warming, and in the case of water bodies, water acidification (Doney, 2006). Both of these are destructive processes that are likely to affect the macro-faunal and -floral community structure and dynamics as well as extinction of species. To date, the effect of cli- mate change on microbial species distribution and abundances has not been shown. However, considering the crucial contribution of microbial metabolism to ecosystem processes, a shift in microbial population dynamics is likely to have severe implications.
- 19. Jiwaji et al. 4 | Microorganisms both consume and produce major greenhouse gases, thus they have central roles both as effectors and monitors of global climate change (DeLong, 2007). The effect of climate change on microorganisms is easiest to monitor in the Antarctica, the coldest, windi- est, driest and most isolated continent on Earth. Due to the environment, Antarctic food webs are relatively simple, and the absence of insect and mammalian herbivores means that most energy and biomass is channelled into a detritus trophic pathway that is dominated by microorganisms (Davis, 1981). Thus, soil microorganisms have a disproportionate importance in nutrient cycling and other ecosystem processes in the ice-free ter- restrial Antarctic ecosystems. The simplicity of the Antarctic ecosystem makes it particularly vulner- able to environmental perturbations like global warming and the Antarctic Peninsula is among the most rapidly warming regions on the planet. This has been demonstrated by significant increases in the abundance of fungi and bacteria and particu- larly in the Alphaproteobacteria-to-Acidobacteria ratio, which is indicative of higher soil nutrient availability (Thomson et al., 2010). The observed shifts are consistent with increased turnover of carbon and nitrogen in the soil upon warming of the environment (Thomson et al., 2010; Yergeau et al., 2012). In the oceans, phytoplankton can also be used as monitors of climate change. These eukaryotic microorganisms influence the biologi- cal pump which cycles carbon, removing it from theupperoceanandtransportingittothedeepsea. Falkowski and Oliver (2007) used the behaviour of phytoplankton to predict the potential effects of climate change on the microorganism commu- nity structure. As these microorganisms are major participants in oceanic primary production, any changes to the phytoplankton community have knock-on effects and implications for fishery pro- ductivity as well as global ocean ecology. The impact of microorganisms in industry From a health and economic point of view, microorganisms are of great interest and they may impact our lives both positively and nega- tively. For example, microbes may be pathogenic and can cause serious diseases (Karlen, 1995). Many microbial species are responsible for the degradation of food produce and can result in significant economic losses. On the positive side, for example, microorganisms are indispensable for their symbiotic roles in the human digestive tract (Eckburg et al., 2005). In addition, they can be harnessed for the production of fuels, chemical compounds, animal feeds, human food, antibiot- ics and pharmaceuticals (Zhu et al., 2012). Microorganisms also represent a vast and dynamic reservoir of genetic variability thus the harnessing of this genetic diversity has allowed scientists to unlock the economic potential of microorganisms. Some of the important meta- bolic processes that microorganisms are involved in include: Nitrogen fixation Nitrogen fixation, the assimilation of atmospheric nitrogen (N2 ) into ammonia, is an economically and environmentally important natural microbial function, which yields hundreds of tonnes of bio- logically available nitrogen and is worth billions of dollars to global agriculture (Herridge et al., 2008). Bioremediation Bioremediation, the use of microorganisms to sequester or remove pollutants, is increasing both as a concept and as an economically viable appli- cation. Non-biological processes are estimated to cost ten-fold more to remediate known hazard- ous waste sites whereas bioremediation would cost less and could occur in the same time frame. This is a developing area and has the potential of becoming economically important. For example, hydrocarbon cold seeps and ecologically devas- tating oil spills have led to the development of marine microorganisms that will utilize alkanes as carbon sources. These bacteria and their enzymes have a potential role in bioremediation and oil processing (Xu et al., 2008; Wasmund et al., 2009; Augustinovic et al., 2012). Bioprospecting Bioprospecting, defined as the discovery and subsequent commercialization of useful products from environmental isolates, has been central in
- 20. Microbes: An Unseen Majority Around Us | 5 the search for novel pharmaceuticals and com- pounds of industrial importance (Dionisi et al., 2012). More than 75% of all antibacterial com- pounds and approximately 50% of all anticancer compounds that are in use clinically are either natural products or derivatives thereof (Newman and Cragg, 2007). The percentage of biologically active natural products is much higher than that of synthetic compounds primarily due to the fact that natural products have evolved bioactivity in a biological context (Firn and Jones, 2003). Since Alexander Fleming’s observation of the antibacterial activity of penicillin in 1928, and its subsequent application to treating bacterial infec- tions, a large number of microbe-derived natural products have been used in the pharmaceutical sector for a diverse range of medical applications. This includes biologically active compounds like the antibiotics cephalosporins, tetracyclines, ami- noglycosides, rifamycins and chloramphenicol, the mycotoxin asperlicin, the immunosuppres- sant cyclosporine and the cholesterol-lowering agent lovastatin (Pan et al., 2010). In addition to the production of pharmaceutically important metabolites, microorganisms also represent a rich source of enzymes with applications as biocatalysts in the production of a wide range of industrially useful compounds. While the pro- duction of these compounds in most instances can be achieved via chemical processes, bio- catalysts are an attractive alternative due to the lower temperatures and neutral pH at which they function, as well as the production of fewer toxic by-products (Azerad, 1995; Koeller and Wong, 2001). Furthermore, biocatalysts often exhibit specific regioselectivity thereby negating the requirement of chemical synthesis pathways for substrate functional-group protection which involves addition of several steps in the synthetic pathway to block and unblock substituents (Schmid et al., 2001). Research into interesting enzymes with potential industrial applications, for example psychrophilic enzymes, is incredibly competitive and lucrative (Struvay and Feller, 2012; Feller, 2013). The range of enzymes with unique abilities includes lipases (de Pascale et al., 2008; Jeon et al., 2009), cellulases (Ekborg et al., 2007; Shan- mughapriya et al., 2009), chitinases (Cottrell et al., 1999; Hobel et al., 2005), proteases, alkane hydroxylase genes (Xu et al., 2008; Wasmund et al., 2009), esterases with high tolerances for salt, organic solvents, cold and high pressure (Auriliaet al., 2008; Chu et al., 2008), and metalloproteases with high temperature optima (Lee et al., 2007). Lipases have wide-ranging applications in the food, detergent, and pharmaceutical industries (de Pascale et al., 2008; Jeon et al., 2009). Cel- lulases have been the subject of intense study because of their role in the generation of biofuels from renewable cellulosic substrates (Ohkuma, 2003; Ekborg et al., 2007; Shanmughapriya et al., 2009). Esterases have great commercial value because of their use in industrial biotransforma- tions (Aurilia et al., 2008; Chu et al., 2008). Some of these enzymes have been isolated using sequence-based approaches by screening for genes with known bioactivity. These methods rely on similarity to known protein families. Others have been isolated using functional screens for enzyme activity (Ferrer et al., 2005). The bio- technological and industrial microbiological potential for new natural products and processes arising from mining microbial diversity is set to increase in the coming years, especially with the application of newly developed techniques, and in particular high-throughput robotic screening technology (Bornscheuer et al., 2012). Assessing microbial populations Traditionally, study of microorganisms has been conducted by staining to visualize the organisms, microscopy and subsequent image analysis. At first, stains were used to quantify the abundance of microorganisms and their biomass (Francisco et al., 1973; Porter and Feig, 1980). Visualization of stained cells was then enhanced by powerful image analysis (Psenner, 1990). More recently, stains have been developed to visualize microor- ganisms like viruses (Ortmann and Suttle, 2009) and to quantify aspects of microbial metabolism (including respiration and enzyme degradation). In addition, radioisotopes (Simon and Azam, 1989) and fluorescent stains (Cotner et al., 2001) have been applied to the measurement of bacterial growth rates.
- 21. Jiwaji et al. 6 | In addition to direct visualization of micro- bial populations, microorganisms have been grown under laboratory conditions, often as pure cultures, and been used for sequence or function-based biological studies. Advances in environmental chemistry (including stable iso- topes, ultrafiltration, measurement of dissolved organic carbon (DOC) and dissolved inorganic carbon (DIC), high performance liquid chro- matography (HPLC) and mass spectrometry (MS) have also enhanced our ability to study microorganisms and their interaction with their environment (Cotner and Biddanda, 2002; Han- delsman, 2004). Up until the 1980s, identification of micro- organisms was achieved by culturing individual isolates in the laboratory, viewing the organisms microscopically and subjecting these cultures to a variety of biochemical tests. While much valu- able data has been generated by this approach, there are several major limitations. Firstly, rep- lication of the complex environment in which microorganisms exist in a laboratory setting is often extremely difficult and it is estimated that less than 0.1% of microorganisms are currently culturable (Stres, 2007). Furthermore, these approaches are time-consuming, labour-intensive and often subjective. Added to this, microorgan- isms that are abundant and/or those that can be cultured under some environmental conditions may change into dormant or possibly uncultur- able forms under other conditions (Hattori et al., 1997). The severity of the problem of relying on culture-dependent studies has been highlighted by cultivation-independent surveys where a major discrepancy can be observed between viable plate count technology and direct methods such as epifluorescence microscopic counts and ssRNA phylogenetic analysis. For example, the observation that marine bacterioplankton are infected by huge viral numbers (tens of billions of phage per litre) and that this phage predation is remarkably specific was undetected primarily because of a lack of representative pure cultures (DeLong, 2007). Viruses, as technically ‘non-living’ entities, cannot be studied with the same methods used to study other microorganisms. Classical methods used for the identification of viruses include in vitro viral amplification followed by cell culture followed by visual observation of cytopathic effects. Virus discovery can also be based on specific nucleic acid hybridization and antigenic cross-reactivity, for example on DNA or antibody microarrays. This can be followed by visualiza- tion using electron microscopy or by testing for immunological cross-reactivity using panels of sera, which is a powerful and rapid method for the identification of the unknown viral agents. Tenta- tive identification of the virus allows the use of more specific molecular approaches like the use of PCR with primers that target the likely viral group for definitive genetic characterization (Delwart, 2007; Mokili et al., 2012). The application of molecular methods bypass the drawbacks presented by culture-dependent methodologies and have led to an improved and deeper understanding of the microbial compo- nents of ecosystems. These new approaches have made apparent our lack of knowledge about envi- ronmental biodiversity. It is estimated that there are ca. 1.5 million taxa that have been described at the species level (de Meeus and Renaud, 2002) yet this represents only a small proportion of the estimated diversity. Currently, there is also a large discrepancy between the microorganism diversity that we detect in biological samples and what is actually present. Molecular ecology studies sug- gest that ca. 1–5% of microbial species have been isolated (Floyd et al., 2005; Hughes Martiny et al., 2006).Andmycologistsestimatethatthereare1.5 million species of fungi despite the fact that only 72,000 species have been isolated or described (Hawkswerth, 1997). Molecular approaches to investigating micro- bial processes can focus either on individual isolates or on the microbial population as a whole within an ecosystem. Either way, a key aim in the molecular biological study of microorgan- isms is to determine the genetic information present within the cell followed by correlation of the encoding genetic material to the complex biological processes which occur within the cell. Bioinformatics analyses are a crucial require- ment, not only in curating and analysing large scale sequence data but also in bridging the gap between genetic code and the encoded function- ality of genes.
- 22. Microbes: An Unseen Majority Around Us | 7 DNA sequence determination Up until recently, sequence analysis of DNA has been achieved by Sanger sequencing. In this method, DNA polymerase-dependent synthesis of a complementary DNA strand occurs in the presence of both natural 2′-deoxynucleotides (dNTPs)and2′,3′-dideoxynucleotides(ddNTPs), which function as irreversible terminators of syn- thesis (Sanger et al., 1977). The DNA synthesis reaction is terminated at random whenever a ddNTP is added to a growing oligonucleotide chain. This results in products of varying lengths with an appropriate ddNTP at the 3′ terminus. These truncated products are separated based on size, using capillary electrophoresis, and the ter- minal ddNTPs used to reveal the DNA sequence of the DNA template strand. This technology has been in use since 1977 and is well established. However, in addition to being lengthy and labour intensive, this methodology suffers from the limi- tation that only one sequence can be generated per capillary which in turn impacts on the amount of genetic information that can be generated. Furthermore, amplification of DNA fragments is required before they can be sequenced. This amplification can be achieved either in vivo, by clonal amplification, or in vitro by PCR, and con- sequently is susceptible to both host-related and PCR biases (Hall, 2007). To deal with the shortcomings of Sanger sequencing, faster, cheaper, and simpler methods for sequencing that bypass the cloning bias and time- and labour-intensive nature of the Sanger method have been developed. These new tech- nologies differ in their approach to generating sequence data, the average read length generated, and the error rate distribution (Shendure and Ji, 2008) thus the appropriate sequencing method needs to be selected based on the application. A brief description of three of these sequencing technologies, namely pyrosequencing, reversible terminator sequencing chemistry and ligation based sequencing chemistry will be covered in this chapter. Detailed information on sequencing techniques can be found in Chapter 2. Pyrosequencing is a sequencing-by-synthesis technique whereby enzyme driven biochemical reactions and chemiluminescence are used to generate sequence data. In this method, the DNA template is sheared and the resultant DNA frag- ments immobilized on beads at a ratio such that a single DNA molecule is immobilized per bead. Individual DNA fragment-carrying beads are then capturedintoseparateemulsiondropletsandthese droplets act as individual amplification chambers which are capable of producing 107 clonal copies of a unique DNA template per bead (Margulies et al., 2005). Each template-containing bead is then transferred into a well of a picotitre plate along with smaller packing beads containing the immo- bilized enzymes ATP sulfurylase and luciferase. The use of the picotitre plate allows hundreds of thousands of pyrosequencing reactions to be car- ried out in parallel, thus increasing the sequencing throughput (Margulies et al., 2005). Once the clonally amplified bead-immobilized DNA frag- ments have been deposited into the picotitre plate wells, the picotitre plate is flooded with dNTP solutions in a sequential manner. As a result of the action of the polymerase, inorganic pyrophos- phate(PPi)isreleasedwheneveracomplementary nucleotide is incorporated into the growing DNA strand. This PPi is converted to ATP by the sul- furylase enzyme and the luciferase in turn utilizes the ATP to convert luciferin to oxyluciferin with light as a by-product. The number of bases incor- porated into the extending DNA strand is directly proportional to the amount of pyrophosphate released which in turn is reflected in the amount of light generated. This chemiluminescent signal is detected by an extremely sensitive camera and interpreted as sequence data (Margulies et al., 2005; Rothberg and Leamon, 2008). Pyrose- quencing is valuable for sequencing regions of DNA that are technically difficult due to strong secondary structure or high GC content as well as for sequencing regions that are resistant to cloning in Escherichia coli (Goldberg et al., 2006). One of the drawbacks of pyrosequencing is the reduced reading accuracy over homopolymeric stretches of identical nucleotides (Ansorge, 2009). The 454 GS FLX+ Platform, which utilizes pyrosequencing chemistry and is currently mar- keted by Roche Applied Science, is capable of generating 700Mb of sequence with a consensus accuracy of 99,997% and a read length of up to 1000bp (http://454.com/products/gs-flx-sys- tem/index.asp, April 2013). The data generated
- 23. Jiwaji et al. 8 | on the 454 platform correlates well with other well established gene expression profiling technologies including microarrays (Torres et al., 2008). An alternative to pyrosequencing is the reversi- ble terminator chemistry (Bennett, 2004; Bennett et al., 2005; Bentley, 2006) which involves what is termed solid-phase bridge amplification of single- molecule DNA templates. Here, one terminal of a single DNA molecule hybridizes to a comple- mentary adaptor which is immobilized on a solid surface, known as a flowcell. With the adapter functioning as a primer, the polymerase extends the DNA fragment thereby generating a copy of the original template. This copy is attached to the flowcell via the adaptor/primer which is now incorporated into the DNA strand. Following denaturation, the flowcell is rinsed to remove the original template leaving the immobilized DNA copy behind. This DNA fragment then bends over and the free terminal end hybridizes to a nearby complementary adaptor forming a DNA ‘bridge’. Once again, the polymerase extends the adaptor/ primer generating a new copy of the DNA which is also attached to the flowcell. This process is repeated until a cluster of ~1000 clonal copies of a single template molecule is formed. After ampli- fication, a flowcell with approximately 40million clusters of DNA amplicons is then subjected to a DNA sequencing-by-synthesis methodology that uses a custom DNA polymerase that can incor- porate specialized reversible terminators with removable fluorescent moieties into growing oli- gonucleotide chains. The terminators are labelled with fluorophores of four different wavelengths to distinguish between the different nucleotide bases and the sequence of the template in each cluster is determined by detecting the wavelength generated at each successive nucleotide addition step. While this methodology is more effective at sequencing homopolymeric stretches than pyrosequencing, the sequence reads are shorter (Bennett et al., 2005; Bentley, 2006; Ansorge, 2009). Also, the custom DNA polymerases and the use of the reversible terminators do result in a higher number of substitution errors (Hutchison, 2007). The HiSeq 2500/1500 platform, which utilizes reversible terminator chemistry, is cur- rently marketed by Illumina. Depending on the run mode (i.e. Rapid-Run Mode or High-Output Run Mode), the HiSeq generates between 95 to 600Gb of sequence with read lengths of between 35 and 150bp (http://www.illumina.com/sys- tems/hiseq_2500_1500.ilmn, April 2013). The third approach to high-throughput sequencing utilizes ligation based chemistry. This approach differs from other next generation tech- nologies in that it is based on hybridization and ligation of fluorescently labelled oligonucleotides rather than polymerase-dependent incorporation of nucleotides. For this approach, an emulsion PCR single-molecule amplification step similar to the one used in the pyrosequencing technique is conducted. The amplification products are then transferred onto a glass surface and sequence analysis occurs via sequential rounds of hybridi- zation and ligation with 16 oligonucleotide octamers labelled with four different fluorescent dyes. After addition, ligation, and fluorescent detection of the complementary octamer, the last two nucleotides with the fluorescent dye attached are cleaved off and the next octamer flowed over the immobilized template DNA. Once the entire length of the DNA fragment has been covered, it is stripped and the ligation of the octamers begins anew from one nucleotide back (n–1) from where the previous hybridizations occurred. As a result, each position on the template is effectively probed twice, and the identity of the nucleotide is deter- mined by analysing the colour that results from two successive ligation reactions. Most impor- tantly, this two-base encoding scheme allows the differentiation between a sequencing error and a sequence polymorphism because an error would be detected in one particular ligation reaction but a polymorphism would be detected in both reactions. This methodology is carried out on the supported oligonucleotide ligation and detection system (SOLiD) supplied by Applied Biosystems. The SOLiD 4™ instrument is capable of generating 80–100Gb of sequence data per run in 35–50bp reads per 8 to 13 day sequencing run (http:// www.appliedbiosystems.com/absite/us/en/ home/applications-technologies/solid-next-gen- eration-sequencing/next-generation-systems/ solid-4-system.html, April 2013). The availability of these high-throughput sequencing technologies has democratized genomics by substantially reducing the cost of
- 24. Microbes: An Unseen Majority Around Us | 9 the technology. In addition, they have removed the need for in vivo cloning by clonal amplifica- tion of spatially separated single molecules using either emulsion PCR or bridge amplification on a solid surface. As these methodologies use single molecule templates, they allow for the detection of heterogeneity in a DNA sample thus provid- ing a powerful advantage over Sanger sequencing (Bentley, 2006; Thomas et al., 2006). Bioinformatics analysis of sequence data Part of bioinformatics research involves the man- agement and analysis of large scale sequence data that has been generated, and is a rapidly growing field of science that incorporates aspects of biol- ogy, mathematics and computer science. Once sequence data have been generated, bioinformat- ics analysis is required as explained in detail in Chapters2,3and4.Dependingontheapplication, input template, and platform utilized, this initial processing would include removal of substandard reads and the alignment of reads into contigs (in the case of whole genome sequencing). The final role played by bioinformaticists is the curation of the huge datasets generated by next generation sequencing technologies. Determination of the nucleotide sequence of a target organism or population of organisms’ genetic material on its own is relatively uninform- ative. Defining how this nucleotide sequence is responsible for structural and metabolic function- ality is more important. Bioinformatics forms the bridge between the sequence data and the biologi- cal functioning in an organism/organisms. Once the nucleotide sequence has been determined, the first step in the bioinformatics analysis of the sequence is gene prediction by detecting potential open reading frames (ORFs). This is achieved by identifying conserved sequences responsible for the initiation and termination of transcription as well as the site for the initiation of translation. Once a potential ORF is identified, the next step is to annotate the putative gene by assigning a func- tion to the sequence. By comparing the encoded queryproteinsequencewithadatabaseofproteins with known functions, putative proteins with suf- ficient homology to the known proteins can then be assigned the corresponding function. If the query protein sequence does not show high levels of similarity with known proteins, annotation by function can be carried out whereby the putative proteins domains can be assigned a function. For example, a particular arrangement of hydrophobic regions within a protein may indicate a membrane protein. While assignment of function based on similarity to other proteins does provide an extremely valuable starting point when correlat- ing DNA sequence to cellular metabolism, it is important to keep in mind that subsequent bio- logical validation is required for confirmation. Application of molecular approaches to the study of microorganisms The genomes of living organisms are essentially barcodes and contain sufficient information to both identify the microorganism as well as outline its physiological functionality (Blaxter, 2003; Blaxter and Floyd, 2003). Genomes contain areas of high and low identity with the differences between the corresponding genes typically clus- tered in sections such as the third (wobble) bases of codons, intronic, and intergenic DNA. As sig- nificant stretches of the genome are maintained by selection to be identical or near-identical between members within a taxon, but which vary between taxa, these segments can be applied to both iden- tification and taxonomy. In addition, as these sequences evolve, they represent both specific and systematic data (Woese, 1987). This makes sequence-based methods incredibly powerful, and this field has revolutionized the ways that we both classify and study microorganisms in their ecosystems, as well as how we screen for novel products and processes. Whole genome sequencing The first microbial genome to be sequenced to completion was that of the human pathogen Hae- mophilus influenzae (Fleischmann et al., 1995). In the same year, the smallest genome of a free-living microorganism, the pathogenic microorganism Mycoplasmagenitalium,wasalsosequenced(Fraser et al., 1995). Since then, the complete sequences of more 4328 microbial genomes have been
- 25. Jiwaji et al. 10 | elucidated (complete and permanent draft) of which 187 are archaeal, 3958 are prokaryotes and 183 are eukaryotic (www.genomesonline.org, April 2013). Initially, genome sequence projects focused on sequencing pathogenic bacteria, now biotechnological consortia are catching up: cur- rently 47% of all bacterial-sequencing projects deal with microorganisms that have industrial applications, 52% of the projects are focused on sequencing pathogens and approximately 1% have targeted exotic organisms (Bode and Muller, 2005; www.genomesonline.org, May 2013; www. tigr.org, May 2013). Knowledge of the genome of a given organism provides tremendous biological insight into cellu- lar processes that may not be evident when using classical culturing/assay techniques which are limited to the study of phenotypic characteristics under the culture/assay conditions. This allows for the identification of genes encoding for poten- tially economically useful metabolites or proteins which may not be produced or expressed under current culture/assay conditions. Furthermore, by increasing our understanding of cellular pro- cesses and mechanisms of gene regulation within target organisms, the ability to optimize microbial metabolism for enhanced application in industry is increased (Xu et al., 2013). In addition to the information generated when sequencing a single genome, the large number of genome sequences whicharecurrentlypubliclyavailableondatabases allows for comparative studies between genomes of different organisms. Such comparative studies can provide valuable information with respect to the encoded function of genes particularly when genomes of closely related strains with differing phenotypes are compared against one another. In addition to novel species, multiple isolates of the same species are also being sequenced. This is due to the fact that even well-known species such as Escherichia coli show large levels of heterogene- ity between strains. For example, comparison of E. coli genomes sequenced to completion show a discrepancy in size from 4.6 to 5.5Mbp. This means that there are close to one million nucleo- tides worth of sequence data that is present in one strain but absent in another (Binnewies et al., 2006). Comparative genomics is discussed in detail in Chapter 5. Currently, the most widely utilized approach to whole genome sequencing is termed ‘shotgun sequencing’(seeChapter2).Inthistechnique,the genome of a chosen microbe is randomly sheared into millions of DNA fragments which are then sequenced. Owing to the random nature of the DNA shearing, many of these fragments will over- lap in terms of sequence data. By aligning these overlapping fragments against one another, it is possible to assemble a larger contiguous sequence (contig). If there are regions of the genome which are not represented in the sequenced fragment library, this will result in contigs which do not overlap and can therefore not be joined together to form a full length sequence of the genome. In this instance, targeted re-sequencing of the miss- ing region is then done by amplifying this region of the genome using primers specific to the termi- nal sequence in the contig. Once the genome has been assembled, the annotation of the genome is begunwherethestructuralandfunctionalfeatures of the genome are identified using bioinformatics tools (discussed in detail in Chapter 4). Microbial species diversity Microbial species diversity in a given environment is determined by the physico-chemical nature of the environment as well as by changes that are caused by the metabolic activities of the microor- ganisms within the community (see Chapter 6). Ideally, a study designed to analyse the biodiver- sity in an environmental sample would examine all the species within the community as well as the size of these species populations. As more envi- ronments are being sampled and analysed, it is evident that the majority of the microbes have yet to be cultured (Rappe and Giovannoni, 2003). This is driving the need for technologies to cul- tivate and enrich formerly ‘not-yet-culturable’ organisms (Stewart, 2012). This is a laborious and time-consuming process and so techniques that circumvent the cultivation of organisms but still allow the exploration of diversity are being explored. When selecting specific genetic regions for analysis of microbial populations, factors such a ubiquity (the target gene must be present in all species), species sequence conservation (to allow for species identification) and evolution-induced
- 26. Microbes: An Unseen Majority Around Us | 11 interspecies variability (to allow for differentia- tion between species as well as to infer taxonomic relatedness) need to be considered (Petti, 2007). The sequences that have been applied to molecu- lar barcoding include the: 1 nuclear small subunit ribosomal RNA gene (SSU, also known as 16S rRNA in prokary- otes, and 18S rRNA in most eukaryotes); 2 nuclear large-subunit ribosomal RNA gene (LSU, also known as 23S rRNA and 28S rRNA gene); 3 internal-transcribed spacer section of the ribosomal RNA cistron (ITS, separated by the 5S ribosomal RNA gene into ITS1 and ITS2 regions); 4 mitochondrial cytochrome c oxidase 1 (CO1 or COX1) gene; and 5 chloroplast ribulose bisphosphate carboxylase large subunit (rbcL) gene (Blaxter, 2004). Analysis of the 16S rRNA gene is discussed in detail in Chapter 8 and genomic barcoding is covered further in Chapter 10. Each gene target has both its benefits and pitfalls and these have to be weighed up when selecting the most appropriate gene. The useful- ness of each of these genes is determined by (i) the ease with which it can be isolated from the sample, (ii) the level of variation between indi- viduals, (iii) the ease with which the amplified sequences can be aligned and analysed and (iv) the availability and number of other sequences from known and identified specimens. It should be kept in mind that sequence-based analyses can be adversely affected if the DNA has extreme base composition biases, polynucleotide runs or very stable secondary structures thus this needs to be taken into account when the targets are selected for analysis. In addition, with a sequence-based system, it is essential that a repository of sequence information, whether these sequences are housed by annexes to databases like EMBL/GenBank or in a freestanding effort, is available (Hebert et al., 2003). For these targets, the PCR can amplify suf- ficient quantities of DNA for study and there are numerous approaches to analyse these amplified gene targets. For example, randomly amplified polymorphic DNA (RAPD), terminal restriction fragment length polymorphism (TRFLP), or denaturing gradient gel electrophoresis (DGGE) allow for the assessment of microbial diversity as well as comparison of community structure between different ecosystems or over time (Liu et al., 1997). While these approaches are rapid and provide useful information in terms of shifts in microbial population dynamics, their major disadvantage lies in the fact that identification of the individual microbes cannot be carried out. In order to identify the species, analysis of the nucleotide sequence must be done. Up until recently, this has been achieved by Sanger-based sequencing of individual clones representing PCR-amplified genes from a few hundred repre- sentatives in a microbial population. The major limitation of this is that only the dominant genotypes within the target population would be sampled. By contrast, next-generation sequencing technologies generate several thousand sequences per sample which consequently allows for the detection of rare biosphere within a given ecosys- tem community. This rare biosphere is important in terms of ecosystem functioning as they may represent critical components of complex con- sortia or be dormant vestiges of a past ecological setting with the potential to become dominant should environmental conditions shift in favour of their growth. An additional advantage of next- generation sequencing technologies is that, due to the large number of sequences generated, a more accurate representation of the relative abundances of the bacterial phylotypes within a target ecosys- tem can be provided (Sogin et al., 2006; Huse et al., 2008). The first studies of bacterial diversity using culture-independent methods have completely changed our view of the biosphere. Many new prokaryotic divisions have been described based on culture-independent studies (Giovannoni et al., 1990; Schmidt et al., 1991; Barns et al., 1994). While some of the sequences were close to known cultured taxa, others showed deeply divergent lineages with no known cultured members. Many of these newly identified lineages are now known to be widespread in many different environments and are sometimes dominant, be it numerically
- 27. Jiwaji et al. 12 | or ecologically. While it is true that some globally distributed ecosystems have a low bacterial diver- sity (Hentschel et al., 2002), most environments, including some of the most inhospitable regions of our planet, showed an unexpected and deep diversity (Pace, 1997; Rothschild and Mancinelli, 2001; Furlong et al., 2002; Venter et al., 2004). However, one must bear in mind that there are pitfalls in assessing microbial diversity using PCR-based techniques as PCR amplification may introduce a bias with each physical, chemical and biological step resulting in a distorted view of the actual environment. This may be caused by the choice of the primers, the conditions of PCR, potential inhibitors that are present in the reactions and variable amplification efficiencies of target genes between species to name a few poten- tial issues (Wintzingerode et al., 1997). Metagenomics Metagenomics provides a gene-based explora- tion of the microbial community as a whole on the basis of genetic material (DNA or RNA) and it returns high resolution data rich information (Moore et al., 2011). The value of this information has been enhanced by the availability of sequence data on a large number of genomes. This topic is discussed in detail in Chapter 7. Metagenomic analysis involves isolating DNA from an environmental sample and analysing the totality of the DNA. As a consequence, the DNA libraries contain the genetic information of all organisms present at a specific location at the sampling time (Daniel, 2004; Streit et al., 2004; Grant et al., 2006). Previously, this DNA would be cloned into suitable vectors, the clones trans- formed into an appropriate host bacterium, and the resulting transformants screened. The clones could be screened for phylogenetic markers, for conserved genes, for expression of specific traits such as enzyme activity or antibiotic production ortheycouldbesequencedviaSangersequencing. In the case of sequencing data, these sequences can be used to query databases allowing for the inference of phylogeny or the identification of putative functional genes. With the availability of next generation sequencing technologies, metagenomes can be analysed without the need for time-consuming and labour intensive cloning steps. Instead, DNA isolated from environmental samples can be sequenced directly (Tuffin et al., 2009).Thisapproachhasbeensuccessfullyapplied to terrestrial and aquatic environments resulting in the discovery of genes for antibiotics, antibiotic resistance and industrial enzymes (D’Costa et al., 2007; Lammle et al., 2007; Suenaga et al., 2007). Examples of enzymes isolated from microorgan- isms using a functional metagenomics approach include lipases, esterases, amylases, amidases and chitinases (Hardeman et al., 2007; Lee et al., 2007; Chu et al., 2008; Xu et al., 2008). Thus the applica- tion of metagenomics has paved the way for the discovery of new genes, proteins and biochemi- cal pathways (Prakash and Taylor, 2012). This technology has also been very important for the identification of new biocatalysts that have been developed by nature, isolated by bioprospecting and optimized by directed evolution (Fernández- Arrojo et al., 2010; Yeh et al., 2011; de Pascale et al, 2012). Viral metagenomics studies have shown that up to 60% of the sequences in a viral prepara- tion are unique, these virus sequences represent unknown viral species that would be missed by traditional Sanger sequencing approaches but are detected by the application of next generation sequencing technologies (Delwart, 2007; Mokili et al., 2012). The full potential of metagenomics is yet to yet be fully realized. This can be attributed to the inability of some metagenomic clones to produce active enzymes. Also, functional metagenomic approaches rely on E. coli as an expression host for metagenome-encoded proteins. While this may occur for a large number of genes, others from more distantly related organisms may not be expressed because of differences in the gene promoters, or the levels of expression may be det- rimentally affected due to differences in the codon usage. If transcription and translation of foreign genes results in the production of protein, it is possiblethatthelimitationsof E. colitopost-trans- lationally modify or export the protein may result in the lack of detectable active protein. The availa- bility of suitable hosts for heterologous expression remains a barrier to efficient mining of functional metagenomic data (Handelsman, 2004). It is pos- sible to couple functional screening approaches to other approaches like substrate-induced gene
- 28. Microbes: An Unseen Majority Around Us | 13 expression screening (SIGEX) to enhance data mining (Uchiyama et al., 2005; Uchiyama and Watanabe, 2007). SIGEX was initially developed to detect metagenomic clones that expressed catabolic genes of interest in the presence of appropriate substrates. The time-consuming and labour-intensive nature of functional screens can also be alleviated by using robotic instrumenta- tion to screen large clone libraries for functional activities in a high-throughput manner (Kennedy et al., 2008). Metagenomics, in addition to being a powerful technique in itself, can be paired with metatran- scriptomics and metaproteomics to generate complementary datasets for a more thorough analysis of the microorganisms in a community (Ram et al., 2005; Frias-Lopez et al., 2008) includ- ing the human microbial biome (discussed in Chapter 8). Despite the recent exciting research advances involving next generation sequenc- ers, it should be noted that the application of these methods is still in its infancy. Efficient and rigorous data analysis pipelines need to be imple- mented and further studies are required to verify the robustness of these techniques as well as the correlation of these results with those obtained by previous methods. Metatranscriptomics and metaproteomics Adaptive responses are driven by changing levels of transcription in the cell as well as changes in the levels of translation. Metatranscriptom- ics, and by extension metaproteomics, focuses on microbial gene expression within complex natural habitats, allowing for culture-independent whole-genome expression profiling of complex microbial communities (Moran, 2009; Sorek and Cossart, 2010; Gosalbes et al., 2011; Mader et al., 2011). However, mining the ‘transcriptome’ and the ‘proteome’, which represent the collection of transcribed sequences and the translated proteins respectively, poses a significant challenge, particu- larly when it comes to comparing data generated ondifferent‘omics’platforms.Therearebothtech- nical and biological hurdles to overcome. Efficient techniques to isolate total environmental RNA and protein are still being developed; however, the inherent complexities of RNA and proteins means that these techniques lag behind those that have been developed for DNA. The relationship between RNA and protein is complex; thus, it is important to be aware of biases in the techniques, for example the differential lifetimes of mRNA and protein. This requires the analysis of temporal changes in transcript and protein levels. The chal- lenge with transcriptomic and proteomic datasets remains the identification of true mRNA–protein concordance and discordance (Hack, 2004; Fang et al., 2008). For this reason, metatranscriptom- ics and metaproteomics are fields that are still being developed, particularly the ability to study gene expression and protein translation in natural environments, which hold special promise for studying microorganism function in ecosystems. The transcriptome refers to coding RNA (mRNA) and noncoding RNA (including rRNA, tRNA, structural RNA, regulatory RNA and other RNA species). In addition, when discuss- ing RNA species, it is important to differentiate between de novo synthesized RNA (capped pri- mary transcripts) and post-transcriptionally modified (uncapped secondary) transcripts as only the processed RNAs will be represented in the proteome of the microorganism. Under- standing transcriptome dynamics is essential for a clearer understanding of the functional output of the genome and can provide valuable insight into gene expression patterns, gene function and regulation (van Vliet, 2010). Early studies of transcriptional activity in microbial cells relied on the sequence analysis of cDNA libraries. Currently, microarrays are the most widely used technique for studying transcriptomes. Arrays provide specific and relative quantification of gene expression free of the bias that is associated with cloning and they allow high-throughput analysis of relative tran- script levels (Hinton et al., 2004). The technique involves the conversion of cellular RNA into labelled cDNA, which in turn is used for hybridi- zation to short oligonucleotides that represent the coding sequences within a genome. To analyse the transcriptome, annotated genome sequences have been used to construct microarrays that represent the majority of all of the predicted genes in a genome. For example, the transcriptomes of Mycoplasma pneumonia, Halobacterium salinarum,
- 29. Jiwaji et al. 14 | Caulobacter crescentus, Bacillus subtilis, Escherichia coli and Listeria monocytogenes have all been ana- lysed on arrays (Selinger et al., 2000; McGrath et al., 2007; Güell et al., 2009; Koide et al., 2009; Rasmussen et al., 2009; Toledo-Arana et al., 2009; Wurtzel et al., 2009). Recently there have been dramatic advances enabling the extension of DNA microarray tech- nology applications to studying environmental transcriptomics (Parro et al., 2007; Moreno-Paz et al., 2010; Pinto et al., 2011). Microarrays that target the 16S rRNA gene have been used to study the population structure of microbial communi- ties (Brodie et al., 2006). Functional gene arrays, for example the Geochip array (He et al., 2007), have been used to study nitrogen and carbon metabolism in Antarctic and PCB contaminated soils (Leigh et al., 2007; Yergeau et al., 2007) and the cellular processes in microbial biofilms (Duran-Pinedo et al., 2011). Phylogenetic micro- arrays are discussed in detail in Chapter 9. While arrays have been instrumental in developing the current understanding of tran- scriptomes, we have started to reach the limits of the applicability of this technology (Bloom et al., 2009). Arrays, like other hybridization-depend- ent techniques, have a relatively limited dynamic range for the detection of the levels of transcripts due to background, saturation and spot density and quality. Arrays need to include sequences that cover multiple strains as mismatches negatively affect hybridization efficiency and the design of appropriate probes is critical to avoid a high background due to nonspecific- or cross- hybridization. The comparison of transcription levels between experiments is challenging and usually requires complex normalization methods (Hinton et al., 2004). Finally, the sensitivity of scanning instruments determines the quality of data that is collected. As probe design limits detection to sequences that are already known, microarrays are more appropriate when they are applied to characterize ecosystems rather than to discover new genes and functions. Also important is the difficulty in differentiating between de novo synthesized transcripts and modified transcripts. While there are techniques that will allow the differentiation of the two sets of transcripts (Hashimoto et al., 2009), the alternative is to couple array-based transcript analysis to a high- throughput sequencing of cDNA libraries (Hoen et al., 2008), resulting in increased transcriptomic data that has been validated by independent plat- forms (Roh et al., 2010). All next generation sequencing technologies can be used for transcriptome sequencing. In this case,totalRNAisextractedfromtheorganismand converted into cDNA by reverse transcription. Because prokaryotic mRNAs lack the poly(A) tail that is typically used for reverse transcription priming in eukaryotic RNA-seq applications, alternative priming approaches are used. These include random hexamer priming (Passalacqua et al., 2009), oligo(dT) priming from artificially polyadenylated mRNAs (Frias-Lopez et al., 2008) and priming from a specific RNA probe ligated to mRNAs (Wurtzel et al., 2009). Sequencing platforms have been used to detect genes, intron usage and alternative initiation codons in yeast (Nagalakshmi et al., 2008; Sorek et al., 2010; van Vliet, 2010; Mader et al., 2011). The application of these platforms is incredibly powerful. In a recent article, 512 new genes were predicted for the nitrogen-fixing plant symbiont Sinorhizobium meliloti after analysis of its transcriptome (Mao et al., 2008). There are, however, technical challenges with metatranscriptomics approaches, which include the need for the standardization of protocols such that datasets can be compared. With environmen- tal samples, technical choices can dramatically affect results and molecular techniques are prone to artefacts and biases no matter what platforms are chosen for sample analysis and data collection (Raz et al., 2011). The reproducible recovery of mRNA and proteins from natural environments is technically challenging as bacterial mRNA often has a very short half-life and hence can be highly unstable (Deutscher, 2003; Condon, 2007). Furthermore, obtaining sufficient mRNA for replicated studies with environmental samples is often difficult. It is essential that future stud- ies consider how to meet accepted standards for independent replication, as this will allow robust statistical analyses of changes or differences in expression patterns (Moran, 2009). In these cases it is necessary first to amplify the RNA. Cao et al. (2010) performed a comparison of several
- 30. Microbes: An Unseen Majority Around Us | 15 methodscapableofamplifyingRNAfortranscrip- tome analyses and found that the polyadenylation of bacterial RNAs and subsequent oligo-dT prim- ing for amplification was sensitive and specific for the measurement of differential gene expression as well as metatranscriptome analyses. Bacterial and archaeal mRNAs are generally unstable and typically do not carry polyA tails thus methods for the specific capture of eukary- otic cDNAs are not applicable (Deutscher, 2003; Wang et al., 2009). Extraction of microbial RNA results in co-extraction of the more abundant and stable rRNAs (which represents more than 80% of total prokaryotic RNA) and tRNAs (Liu et al., 2009; Yoder-Himes et al., 2009). This can lead to low yields of expressed gene sequences in a large- scale sequencing run, potentially as low as 10% (Moran, 2009). Many sequencing protocols using RNA as the template rely on the removal of rRNA by subtractive hybridization (Chen and Duan, 2011; Liu and Camilli, 2011), or exonuclease treatment (Sharma et al., 2010) before the prepa- ration of cDNA libraries. It has to be borne in mind that any treatments, including rRNA deple- tion methods, will probably introduce unknown biases so alternative approaches are being consid- ered, for example the NSR-Seq (not so random) that uses computationally designed hexamers to selectively enrich for mRNA transcripts during cDNA synthesis (Armour et al., 2009; Hirakawa et al., 2011). Metaproteomics promises to be an exciting techniquethatiscomplementarytometatranscrip- tomics (Maron et al., 2007; Siggins et al., 2011). Metaproteomic data provides a less-resolved view of instantaneous regulatory responses than metatranscriptomics, but will provide a better link to metabolic function. Successful application of this protein-based technique relies on effective recovery of proteins from samples. Ideally, a pro- cedure for recovering microbial proteins should allow highly efficient protein recovery to obtain a protein pool that is (1) of sufficient purity for analysis using the biochemical methods available, and (2) representative of the total proteins within the natural microbial community (Vieites et al., 2009). This approach was first used to analyse the community proteome in a natural biofilm growing inside the Richmond Mine at Iron Mountain, northern California, in the USA, and successfully combined mass spectrometry proteomics with community genomic analyses to yield rich and robust data (Ram et al., 2005). This approach is becoming increasingly important in under- standing the role of the microbiota in ecosystem functioning in the environment (Abram et al., 2011; Wang et al., 2011). In a more recent study, Kolmeder et al. (2012) used the same approach to investigate the stability and function of the human intestinal microbiota. The profiling of biological samples for biochemical reaction products, or so-called metabolome, serves to elucidate the main meta- bolic pathways and metabolic bottlenecks and, in the context of microbial communities, it may be helpful to access and track the complex metabolic interactions between microorganisms. Being a post-genomics tool, metabolomics is a young and vibrant field of research still in its growth phase (Motti, 2012). At present, the technical issues of protein extraction, separation, and identification make metaproteomics more challenging than metatranscriptomics. There is much still to be done in the development of the metaproteomic technologies until they can be routinely applied to analysis of the metaproteome in environmental samples. As sequencing information is accumulated, bio- informatics techniques are also being applied to turning this information into knowledge such that scientistscanassessmicrobesatthemolecularand the functional level. This is assisted by the devel- opment of specific computational biology tools to assess the phylogenetic relationships and meta- bolic functions on the basis of the comparison of gene sequences (Prakash and Taylor, 2012). Many technical challenges remain in metagen- omic, metatranscriptomic and metaproteomic approaches, including extraction of the sample material (be it DNA, RNA, or protein), mRNA instability, low abundance, and low proportion of mRNA in total RNA (Wooley and Ye, 2009; Carvalhais et al., 2012; Muth et al., 2012). Micro- fluidics, the manipulation of small volumes and by
- 31. Jiwaji et al. 16 | extrapolation,smallersamples,willenhanceallthe high throughput meta techniques. The combina- tion of high throughput screening, microfluidics, and dilution-to-extinction techniques can make accessible previously unculturable species such that they can be characterized and their biotech- nological potential evaluated. As these are still relatively new approaches, solutions to these tech- nical problems will be developed as researchers are required to address them. In addition, there are the bioinformatic chal- lenges (Schneider and Orchard, 2011). With the advent of the ‘omics’ era, metagenomic, metatran- scriptomic, and metaproteomic technologies are producing more and more large datasets from the analysis of environmental samples. Analysis resources and tools have been developed in an attempt to maximize the meaningful data that can be obtained from these vast datasets. This requires the available capacity in the public databases for data storage. Currently, there are a few public databases that store annotated microbial genomes and allow various types of searches and sequence analyses. However, the number of data reposito- ries for environmental sequences is limited. And although GenBank serves as the main repository for all public sequences, the annotation quality for environmental data is poor and the options for comparative analyses are limited (Vieites et al., 2009). The lack of reference sequences and genomes presents an additional difficulty when approach- ing metagenomic data however, the increase in available data and analysis tools should stead- ily reduce this problem. Also, the inclusion of comprehensive sampling and processing meth- odologies will support the contextualization of sequence data thereby improving interpretation of data obtained (Kennedy et al., 2010). Another important and difficult challenge for all ‘meta’ approaches is that only a small percentage of the vast number of ecologically important genes have been correctly annotated and sequence datasets often contain only the most abundant genes from a very limited number of natural microbial com- munities. Furthermore, many sequences cannot be confidently assigned a function as they have no close matches in existing databases. For example, in a metatranscriptomics study by Poretsky et al. (2009), only 33% of the putative protein-encod- ing sequences exhibited homology to annotated proteins in the NCBI RefSeq database, only 24% had matches to a KEGG pathway, and only 16% to a COG category. It is essential to have accurate sequence determinations to be able to map cDNA readsontothegenomeandtoremovepoor-quality sequences (Marioni et al., 2008; Jiang and Wong, 2009; Oshlack and Wakefield, 2009) which makes this a limitation to detailed analysis of ‘meta’ data. While the larger datasets will allow more accurate determination of transcript levels and associated statistics, they also increase the risk of large volumes of meaningless data. Visualization, analysis and interpretation of these large datasets require significant levels of expertise and some- times appropriate programming skills. There are some bioinformatics tools available, for example ARTEMIS (Carver et al., 2008), LASERGENE (DNAstar) and CAMERA (Seshardi et al., 2007), however continued and coordinated effort will be required to develop new tools for analysis of burgeoning datasets as these large datasets need to be critically analysed. In particular, databases and software tools are essential and their further development is needed to deal with the grow- ing bottleneck in the analysis of metagenomic, metatranscriptomic and metaproteomic data. Concluding remarks Biodiversity is defined as ‘all hereditarily based variation at all levels of organization, from the genes within a single local population or species, to the species composing all or part of a local com- munity,andfinallytothecommunitiesthemselves that compose the living parts of the multifarious ecosystems of the world’ (Wilson, 1997). By this definition, microbial diversity represents the genetic composition of the microorganisms and the environment or habitat in which they are found, as well as their ecological or functional role within the ecosystem. To better understand the environmental biodi- versity, new approaches are being developed that are likely to have a major impact on biodiscovery of novel enzymes from microorganisms that are difficult to culture. These new methodologies include single cell analysis (Fritzsch et al., 2012),
- 32. Microbes: An Unseen Majority Around Us | 17 high-throughput nanoscale sequencing (Wanunu, 2012) and sequencing single molecules (Kumar et al., 2012). Whilethemetagenomicsapproachesdescribed above are useful for exploiting the biochemistry of microorganisms in a community, they are how- ever unable to access the metabolic capabilities associated with specific microorganisms within the consortia; given that they largely rely more on a ‘take all’ approach (Kennedy et al., 2010). In this era of generation of large datasets (i.e. metagen- omics is providing a gene-based exploration of the community as a whole), the classical techniques still have their place as the cultivation and analysis of individual members of this community will continue to be the mainstay for testing metabolic abilities and more detailed genomic studies. Emerging DNA, RNA, and protein sequenc- ing technologies are exponentially increasing our understanding of the microbial world from the level of a single cell to that of a complex microbial consortia and from environments ranging from deep sea thermal vents to acidic hot springs, permafrost to desert soils, the human mouth to termite guts. The new technologies have resulted in a situation where the limitation is no longer the ability to produce biological data but rather the challenge now is with developing effective tools to analyse the data to generate meaningful information. The demand for improved bioinfor- matics tools has stimulated an extremely active field of science which is continually expanding and enriching our understanding of microorgan- isms. References Abram, F., Enright, A.M., O’Reilly, J., Botting, C.H., Col- lins, G., and O’Flaherty, V. (2011). A metaproteomic approach gives functional insights into anaerobic digestion. J. Appl. Microbiol. 110, 1550–1560. Ansorge, W.J. (2009). Next-generation DNA sequencing techniques. New Biotechnol. 25, 195–203. Armour, C.D., Castle, J.C., Chen, R., Babak, T., Loerch, P., Jackson, S., Shah, J.K., Dey, J., Rohl, C.A., Johnson, J.M., and Raymond, C.K. (2009). Digital transcrip- tome profiling using selective hexamer priming for cDNA synthesis. Nat. Methods 6, 647–649. Augustinovic, Z., Birketveit, O., Clements, K., Freeman, M., Gopi, S., Ishoey, T., Jackson, G., Kubala, G., Larsen, J., Marcotte, B.W.G., et al. (2012). Microbes- oilfield enemies or allies? Oilfield Rev. 24, 4–17. Aurilia, V., Parracino, A., and D’Auria, S. (2008). Microbial carbohydrate esterases in cold adapted environments. Gene 410, 234–240. Azerad, R. (1995). Application of biocatalysis in organic synthesis. Bull. Soc. Chim. Fr. 132, 17–51. Baker-Austin, C., and Dopson, M. (2007). Life in acid: pH homeostasis in acidophiles. Trends Microbiol. 15, 165–171. Barns, S.M., Fundygam, R.E., Jeffries, M.W., and Pace, N.R. (1994). Remarkable archaeal diversity detected in a Yellowstone National Park hot spring environ- ment. Proc. Natl. Acad. Sci. U.S.A. 91, 1609–1613. Bell, T., Newman, J.A., Silverman, B.W., Turner, S.L., and Lilley, A.K. (2005). The contribution of species rich- ness and composition to bacterial services. Nature 436, 1157–1160. Bennett, S. (2004). Solexa Ltd. Pharmacogenomics J. 5, 433–438. Bennett, S.T., Barnes, C., Cox, A., Davies, L., and Brown, C. (2005). Toward the 1,000 dollars human genome. Pharmacogenomics J. 6, 373–382. Bentley, D.R. (2006). Whole-genome re-sequencing, Curr. Opin. Genet. Dev. 16, 545–552. Binnewies, T.T., Motro, Y., Hallin, P.F., Lund, O., Dunn, D., La, T., Hampson, D.J., Bellgard, M., Wassenaar, T.M., and Usserv, D.W. (2006) Ten years of bacterial genome sequencing: comparative-genomics-based discoveries. Funct. Integr. Genomics 6, 165–85. Blaxter, M.L. (2003). Molecular systematics: counting angels with DNA. Nature 421, 122–124. Blaxter, M.L. (2004). The promise of a DNA taxonomy. Phil. Trans. R. Soc. Lond. B 359, 669–679. Blaxter, M.L., and Floyd, R. (2003). Molecular taxonom- ics for biodiversity surveys: already a reality. Trends Ecol. Evol. 18, 268–269. Bloom, J.S., Khan, Z., Kruglyak, L., Singh, M., and Caudy, A.A. (2009). Measuring differential gene expression by short read sequencing: quantitative comparison to 2-channel gene expression microarrays. BMC Genom. 10, 221. Bode, H.B., and Muller, R. (2005). The impact of bacterial genomics on natural product research. Angew. Chem. 44, 6828–6846. Bornscheuer, U.T., Huisman, G.W., Kazlauskas, R.J., Lutz, S., Moore, J.C., and Robins, K. (2012). Engineering the third wave of biocatalysis. Nature 485, 185–194. Brodie, E.L., Desantis, T.Z., Joyner, D.C., Baek, S.M., Larsen, J.T., Andersen, G.L., Hazen, T.C., Richardson, P.M., Herman, D.J., Tokunaga, T.K., Wan, J.M., and Firestone, M.K. (2006). Application of a high-density oligonucleotide microarray approach to study bacterial population dynamics during uranium reduction and reoxidation. Appl. Environ. Microbiol. 72, 6288–6298. Cao, F.L., Liu, H.H., Wang, Y.H., Liu, Y., Zhang, X.Y., Zhao, J.Q., Sun, Y.M., Zhou, J., and Zhang, L. (2010). An optimized RNA amplification method for prokary- otic expression profiling analysis. Appl. Microbiol. Biotechnol. 87, 343–352. Carvalhais, L.C., Dennis, P.G., Tyson, G.W., and Schenk, P.M. (2012). Application of metatranscriptomics to soil environments. J. Microbiol. Methods 91, 246–51.
- 33. Jiwaji et al. 18 | Carver, T., Berriman, M., Tivey, A., Patel, C., Bohme, U., Barrell, B.G., Parkhill, J., and Rajandream, M.A. (2008). Artemis and ACT: viewing, annotating and comparing sequences stored in a relational database. Bioinformatics 24, 2672–2676. Cavigelli, M.A., and Robertson, G.P. (2000). The functional significance of denitrifier community composition in a terrestrial ecosystem. Ecology 81, 1402–1414. Chen, Z., and Duan, X. (2011) Ribosomal RNA depletion for massively parallel bacterial RNA-sequencing appli- cations. Methods Mol. Biol. 733, 93–103. Chu, X., He, H., Guo, C., and Sun, B. (2008). Identifica- tionoftwonovelesterasesfromamarinemetagenomic libraryderivedfromSouthChinaSea.Appl.Microbiol. Biotechnol. 80, 615–625. Condon, C. (2007). Maturation and degradation of RNA in bacteria. Curr. Opin. Microbiol. 10, 271–278. Cotner, J.B., and Biddanda, B.A. (2002). Small Players, Large Role: Microbial Influence on Biogeochemical Processes in Pelagic Aquatic Ecosystems. Ecosystems 5, 105–121. Cotner, J.B., Ogdahl, M.L., and Biddanda, B.A. (2001). Double-stranded DNA measurement in lakes with the fluorescent stain PicoGreen and the application to bacterial bioassays. Aquat. Microb. Ecol. 25, 65–74. Cottrell, M.T., Moore, J.A., and Kirchman, D.L. (1999). Chitinases from uncultured marine microorganisms. Appl. Environ. Microbiol. 65, 2553–2557. Daniel, R. (2004). The soil metagenome – a rich resource for the discovery of novel natural products. Curr. Opin. Biotechnol. 15, 199–204. Davis, R.C. (1981). Structure and function of two Ant- arctic terrestrial moss communities. Ecol. Monogr. 51, 125–143. D’Costa, V.M., Griffiths, E., and Wright, G.D. (2007) Expanding the soil antibiotic resistome: exploring environmental diversity. Curr. Opin. Microbiol. 10, 481–489. DeLong, E.F. (2007). Modern microbial seascapes. Nat. Rev. Microbiol. 5, 755–757. DeLong, E.F., and Karl, D.M. (2005). Genomic perspec- tives in microbial oceanography. Nature 437, 336–342. Delwart, E.L. (2007). Viral metagenomics. Rev. Med. Virol. 17, 115–131. Demain, A.L. (2007). The business of biotechnology. Ind. Biotechnol. 3, 269–283. Desai,C.,Pathak,H.,andMadamwar,D.(2010).Advances in molecular and ‘-omics’ technologies to gauge micro- bial communities and bioremediation at xenobiotic/ anthropogen contaminated sites. Bioresour. Technol. 101, 1558–69. Deutscher, M.P. (2003). Degradation of stable RNA in bacteria. J. Biol. Chem. 278, 45041–45044. De Pascale, D., Cusano, A.M., Autore, F., Parrilli, E., di Prisco, G., Marino, G., and Tutino, M.L. (2008). The cold-active Lip1 lipase from the Antarctic bacterium Pseudoalteromonas haloplanktis TAC125 is a member of a new bacterial lipolytic enzyme family. Extremo- philes 12, 311–323. De Pascale, D., de Santi, C., Fu, J., and Landfald, B. (2012). The microbial diversity of Polar environments is a fertile ground for bioprospecting. Mar. Genomics 8, 15–22. Dionisi, H.M., Lozada, M., and Olivera, N.L. (2012). Bioprospection of marine microorganisms: biotech- nological applications and methods. Rev. Argentina Microbiol. 44, 49–60. Doney, S.C. (2006). The dangers of ocean acidification. Sci. Am. 294, 58–65. Duran-Pinedo, A.E., Paster, B., Teles, R., and Frias-Lopez, J. (2011). Correlation network analysis applied to complex biofilm communities. PLoS One 6, e28438. Eckburg, P.B., Bik, E.M., Bernstein, C.N., Purdom, E., Dethlefsen, L., Sargent, M., Gill, S.R., Nelson, K.E., and Relman, D.A. (2005). Diversity of the human intestinal microbial flora. Science 308, 1635–1638. Ekborg, N.A., Morrill, W., Burgoyne, A.M., Li, L., and Distel, D.L. (2007). CelAB, a multifunctional cellulase encoded by Teredinibacter turnerae T7902T, a cultur- able symbiont isolated from the wood-boring marine bivalve Lyrodus pedicellatus. Appl. Environ. Microbiol. 73, 7785–7788. Fagerbakke, K.M., Heldal, M., and Norland, S. (1996). Content of carbon, nitrogen, oxygen, sulfur and phos- phorus in native aquatic and cultured bacteria. Aquat. Microb. Ecol. 10, 15–27. Falkowski, P.G., and Oliver, M.J. (2007). Mix and match: how climate selects phytoplankton. Nat. Rev. Micro- biol. 5, 813–819. Fang, H., Wang, K., and Zhang, J. (2008). Transcriptome and Proteome Analyses of Drug Interactions with Natural Products. Curr. Drug Metab. 9, 1038–1048. Fang, J., Zhang, L., and Bazylinski, D.A. (2010). Deep-sea piezosphere and piezophiles: geomicrobiology and biogeochemistry. Trends Microbiol. 18, 413–422. Feller, G. (2013). Psychrophilic Enzymes: From Folding to Function and Biotechnology. Scientifica http:// dx.doi.org/10.1155/2013/512840 Fernández-Arrojo, L., Guazzaroni, M.E., López-Cortés, N., Beloqui, A., and Ferrer, M. (2010). Metagenomic era for biocatalyst identification. Curr. Opin. Biotech- nol. 21, 725–33. Ferrer, M., Golyshina, O.V., Chernikova, T.N., Khachane, A.N., Martins Dos Santos, V.A., Yakimov, M.M., Timmis, K.N., and Golyshin, P.N. (2005). Microbial enzymes mined from the Urania deepsea hypersaline anoxic basin. Chem. Biol. 12, 895–904. Firn, R.D., and Jones, C.G. (2003). Natural products – a simple model to explain chemical diversity. Nat. Prod. Rep. 20, 382–391. Fleischmann, R.D., Adams, M.D., White, O., Clayton, R.A., Kirkness, E.F., Kerlavage, A.R., Bult, C.J., Tomb, J.F., Dougherty, B.A., Merrick, J.M., et al. (1995). Whole genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269, 496–512. Floyd, M.M., Tang, J., Kane, M., and Emerson, D. (2005). Captured diversity in a culture collection: case study of the geographic and habitat distribu- tions of environmental isolates held at the American type culture collection. Appl. Environ. Microbiol. 71, 2813–2823. Francisco, D., Mah, R., and Rabin, A. (1973). Acridine orange–epifluorescence technique for counting
- 34. Microbes: An Unseen Majority Around Us | 19 bacteria in natural waters. Trans. Am. Microscop. Soc. 92, 416–21. Fraser, C.M., Gocayne, J.D., White, O., Adams, M.D., Clayton, R.A., Fleischmann, R.D., Bult, C.J., Kerlav- age, A.R., Sutton, G., Kelley, J.M., et al. (1995) The minimal gene complement of Mycoplasma genitalium. Science 270, 397–403. Frias-Lopez, J., Shi, Y., Tyson, G.W., Coleman, M.L., Schuster, S.C., Chisholm, S.W., and DeLong, E.F. (2008). Microbial community gene expression in ocean surface waters. Proc. Natl. Acad. Sci. U.S.A. 105, 3805–3810. Fritzsch, F.S., Dusny, C., Frick, O.,and Schmid, A. (2012). Single-cell analysis in biotechnology, systems biology, and biocatalysis. Annu. Rev. Chem. Biomol. Eng. 3, 129–155. Furlong, M.A., Singleton, D.R., Coleman, D.C., and Whitman, W.B. (2002). Molecular and culture-based analyses of prokaryotic communities from an agricul- tural soil and the burrows and casts of the earthworm Lumbricus rubellus. Appl. Environ. Microbiol. 68, 1265–1279. Giovannoni, S.J., Britschgi, T.B., Moyer, C.L., and Field, K.G. (1990). Genetic diversity in Sargasso Sea bacte- rioplankton. Nature 345, 60–63. Goldberg, S.M., Johnson, J., Busam, D., Feldblyum, T., Ferriera, S., Friedman, R., Halpern, A., Khouri, H., Kravitz, S.A., Lauro, F.M., et al. (2006). A Sanger/ pyrosequencing hybrid approach for the generation of high-quality draft assemblies of marine microbial genomes. Proc. Natl. Acad. Sci. U.S.A. 103, 11240– 11245. Gosalbes, M.J., Durban, A., Pignatelli, M., Abellan, J.J., Jimenez- Hernandez, N., Perez-Cobas, A.E., Latorre, A., and Moya, A. (2011). Metatranscriptomic approach to analyze the functional human gut micro- biota. PLoS One 6, e17447. Grant, S., Grant, W.D., Cowan, D.A., Jones, B.E., Ma, Y., Ventosa, A., and Heaphy, S. (2006). Identification of eukaryotic open reading frames in metagenomic cDNA libraries made from environmental samples. Appl. Environ. Microbiol. 72, 135–43. Güell, M., van Noort, V., Yus, E., Chen, W.-H., Leigh-Bell, J., Michalodimitrakis, K., Yamada, T., Arumugam, M., Doerks, T., Kühner, S., et al. (2009). Transcriptome complexity in a genome-reduced bacteria. Science 326, 1268–1271. Hack,C.J.(2004).Integratedtranscriptomeandproteome data: The challenges ahead. Brief. Funct. Genom. Pro- teom. 3, 212–219. Hall, N. (2007) Advanced sequencing technologies and their wider impact in microbiology. J. Exp. Biol. 210, 1518–1525. Handelsman, J. (2004). Metagenomics: Application of Genomics to Uncultured Microorganisms. Microbiol. Mol. Biol. Rev. 68, 669–685. Hardeman, F., and Sjoling, S. (2007). Metagenomic approach for the isolation of a novel low-tempera- tureactive lipase from uncultured bacteria of marine sediment. FEMS Microbiol. Ecol. 59, 524–534. Hashimoto, S., Qu, W., Ahsan, B., Ogoshi, K., Sasaki, A., Nakatani, Y., Lee, Y., Ogawa, M., Ametani, A., Suzuki, Y., et al. (2009). High-resolution analysis of the 5′-end transcriptomeusinganextgenerationDNAsequencer. PLoS One 4, e4108. Hattori, T., Mitsui, H., Haga, H., Wakao, N., Shikano, S., Gorlach, K., Kasahara, Y., El-Beltagy, A., and Hattori, R. (1997). Advances in soil microbial ecology and the biodiversity. Antonie van Leeuwenhoek 72, 21–28. Hawkswerth, D.L. (1997). The fascination of fungi: exploring fungal diversity. Mycologist 11, 18–22. He, Z., Gentry, T.J., Schadt, C.W., Wu, L., Liebich, J., Chong, S.C., Huang, Z., Wu, W., Gu, B., Jardine, P., Criddle, C., and Zhou, J. (2007). GeoChip: a compre- hensive microarray for investigating biogeochemical, ecological and environmental processes. ISME J. 1, 67–77. Hebert, P.D.N., Cywinska, A., Ball, S.L., and de Waard, J.R. (2003). Biological identifications through DNA barcodes. Proc. R. Soc. Lond. B 270, 313–321. van der Heijden, M.G.A., Klironomos, J.N., Ursic, M., Moutoglis, P., Streitwolf-Engel, R., Boller, T., Wiem- ken, A., and Sanders, I.R. (1998). Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396, 69–72. Hentschel, U., Hopke, J., Horn, M., Friedrich, A.B., Wagner, M., Hacker, J., and Moore, B.S. (2002). Molecular evidence for a uniform microbial commu- nity in sponges from different oceans. Appl. Environ. Microbiol. 68, 4431–4440. Herridge, D.F., Peoples, M.B., and Boddey, R.M. (2008). Global inputs of biological nitrogen fixation in agricul- tural systems. Plant Soil 311, 1–18. Hibbing, M.E., Fuqua, C., Parsek, M.R., and Peterson, S.B. (2010). Bacterial competition: surviving and thriving in the microbial jungle. Nat. Rev. Microbiol. 8, 15–25. Hinton, J.C., Hautefort, I., Eriksson, S., Thompson, A., and Rhen, M. (2004). Benefits and pitfalls of using micro- arrays to monitor bacterial gene expression during infection. Curr. Opin. Microbiol. 7, 277–282. Hirakawa, H., Oda, Y., Phattarasukol, S., Armour, C.D., Castle, J.C., Raymond, C.K., Lappala, C.R., Schaefer, A.L., Harwood, C.S., and Greenberg, E.P. (2011). Activity of the Rhodopseudomonas palustris p-cou- maroyl-homoserine lactone-responsive transcription factor RpaR. J. Bacteriol. 193, 2598–2607. Hobel, C.F., Hreggvidsson, G.O., Marteinsson, V.T., Bah- rani-Mougeot, F., Einarsson, J.M., and Kristjansson, J.K. (2005). Cloning, expression, and characterization of a highly thermostable family 18 chitinase from Rho- dothermus marinus. Extremophiles 9, 53–64. Hoen, P.A.T., Ariyurek, Y., Thygesen, H.H., Vreugdenhil, E., Vossen, R.H., de Menezes, R.X., Boer, J.M., van Ommen, G.J., and den Dunnen, J.T. (2008). Deep sequencing-based expression analysis shows major advances in robustness, resolution and inter-lab port- ability over five microarray platforms. Nucleic Acids Res. 36, e141. Horz, H.-P., Barbrook, A., Field, C.B., and Bohannan, B.J.M. (2004). Ammonia-oxidizing bacteria respond to multifactorial global change. Proc. Natl Acad. Sci. U.S.A. 101, 15136–15141. Hughes Martiny, J.B., Bohannan, B.J.M., Brown, J.H., Col- well, R.K., Fuhrman, J.A., Green, J.L., Horner-Devine,
- 35. Jiwaji et al. 20 | M.C., Kane, M., Adams Krumins, J., Kuske, C.R., et al. (2006). Microbial biogeography: putting microorgan- isms on the map. Nat. Rev. Microbiol. 4, 102–112. Huse, S.M., Dethlefsen, L., Huber, J.A., Welch, D.M., Relman, D.A., and Sogin, M.L. (2008). Exploring microbial diversity and taxonomy Using SSU rRNA Hypervariable Tag Sequencing. PLoS Genet. 4, e1000255. Hutchison III, C.A. (2007). DNA sequencing: bench to bedside and beyond. Nucleic Acids Res. 35, 6227– 6237. Jeon, J.H., Kim, J.T., Kim, Y.J., Kim, H.K., Lee, H.S., Kang, S.G., Kim, S.J., and Lee, J.H. (2009). Cloning and char- acterization of a new cold-active lipase from a deep-sea sediment metagenome. Appl. Microbiol. Biotechnol. 81, 865–874. Jiang, H., and Wong, W.H. (2009). Statistical inferences for isoform expression in RNA-Seq. Bioinf. 25, 1026–1032. Karl, D.M. (2007). Microbial oceanography: Paradigms, processes and promise. Nat. Rev. Microbiol. 5, 759–769. Karlen, A. (1995). Man and Microbes: Disease and Plagues in History and Modern Times. (New York: G.P. Putnam’s). Kennedy, J., Marchesi, J.R., and Dobson, A.D. (2008). Marine metagenomics: strategies for the discovery of novel enzymes with biotechnological applications from marine environments. Microb. Cell Fact. 7, 27. Kennedy, J., Flemer, B., Jackson, S.A., Lejon, D.P.H., Mor- rissey, J.P., O’Gara, F., and Dobson, A.D.W. (2010). Marine Metagenomics: New Tools for the Study and Exploitation of Marine Microbial Metabolism. Mar. Drugs 8, 608–628. Koeller, K.M., and Wong, C. (2001). Enzymes for chemi- cal synthesis. Nature 409, 232–240. Koide, T., Reiss, D.J., Bare, J.C., Pang, W.L., Facciotti, M.T., Schmid, A.K., Pan, M., Marzolf, B., Van, P.T., Lo, F.Y., et al. (2009). Prevalence of transcription promoters within archaeal operons and coding sequences. Mol. Syst. Biol. 5, 285. Kolmeder, C.A., de Been, M., Nikkilä, J., Ritamo, I., Mättö, J., Valmu, L., Salojärvi, J., Palva, A., Salonen, A., and de Vos, W.M. (2012). Comparative metaproteomics and diversity analysis of human intestinal microbiota testifies for its temporal stability and expression of core functions. PLoS One 7, e29913. Kumar, S., Tao, C., Chien, M., Hellner, B., Balijepalli, A., Robertson, J.W., Li, Z., Russo, J.J., Reiner, J.E., Kasianowicz, J.J.,and Ju, J. (2012). PEG-labeled nucle- otides and nanopore detection for single molecule DNA sequencing by synthesis. Sci. Rep. 2, 684. Lammle, K., Zipper, H., Breuer, M., Hauer, B., Buta, C., Brunner, H., and Rupp, S. (2007). Identification of novel enzymes with different hydrolytic activities by metagenome expression cloning. J. Biotechnol. 127, 575–592. Lee, D.G., Jeon, J.H., Jang, M.K., Kim, N.Y., Lee, J.H., Lee, J.H., Kim, S.J., Kim, G.D., and Lee, S.H. (2007). Screening and characterization of a novel fibrinolytic metalloprotease from a metagenomic library. Biotech- nol. Lett. 29, 465–472. van Leeuwenhoek, A. (1677) Concerning little animals by him observed in rain-, well-, sea- and snow-water; as also in water wherein pepper had lain infused. Phil. Trans. Royal Soc. London 12, 821–831. Leigh,M.B.,Pellizari,V.H.,Uhlik,O.,Sutka,R.,Rodrigues, J., Ostrom, N.E., Zhou, J., and Tiedje, J.M. (2007). Biphenyl-utilizing bacteria and their functional genes in a pine root zone contaminated with polychlorinated biphenyls (PCBs). ISME J. 1, 134–148. Liu, J.M., and Camilli, A. (2011). Discovery of bacterial sRNAs by high-throughput sequencing. Methods Mol. Biol. 733, 63–79. Liu, J.M., Livny, J., Lawrence, M.S., Kimball, M.D., Waldor, M.K., and Camilli, A. (2009). Experimental discovery of sRNAs in Vibrio cholerae by direct cloning, 5S/ tRNA depletion and parallel sequencing. Nucleic Acids Res. 37, e46. Liu, W.-T., Marsh, T.L., Cheng, H., and Forney, L. (1997). Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16SrRNA. Appl. Environ. Micro- biol. 63, 4516–4522. McGrady-Steed, J., Harris, P.M., and Morin, P.J. (1997). Biodiversity regulates ecosystem predictability. Nature 390, 162–165. McGrath, P.T., Lee, H., Zhang, L., Iniesta, A.A., Hottes, A.K., Tan, M.H., Hillson, N.J., Hu, P., Shapiro, L., and McAdams, H.H. (2007). High-throughput identifica- tion of transcription start sites, conserved promoter motifs and predicted regulons. Nat. Biotech. 25, 584–592. Mader, U., Nicolas, P., Richard, H., Bessieres, P., and Aymerich, S. (2011). Comprehensive identification and quantification of microbial transcriptomes by genome-wide unbiased methods. Curr. Opin. Biotech- nol. 22, 32–41. Mao, C., Evans, C., Jensen, R.V., and Sobral, B.W. (2008). Identification of new genes in Sinorhizobium meliloti using the Genome Sequencer FLX system. BMC Microbiol. 8, 72. Margulies, M., Egholm, M., Altman, W.E., Attiya, S., Bader, J.S., Bemben, L.A., Berka, J., Braverman, M.S., Chen, Y.-J., Chen, Z., et al. (2005). Genome sequencing in microfabricated high-density picolitre reactors. Nature 437, 376–380. Marioni, J.C., Mason, C.E., Mane, S.M., Stephens, M., and Gilad, Y. (2008). RNA-seq: an assessment of technical reproducibility and comparison with gene expression arrays. Genome Res. 18, 1509–1517. Maron, P.A., Ranjard, L., Mougel, C., and Lemanceau, P. (2007). Metaproteomics: a new approach for study- ing functional microbial ecology. Microb. Ecol. 53, 486–93. de Meeus, T., and Renaud, F. (2002). Parasites within the new phylogeny of eukaryotes. Trends Parasitol. 18, 247–251. Mokili, J.L., Rohwer, F., and Dutilh, B.E. (2012). Metagen- omics and future perspectives in virus discovery. Curr. Opin. Virol. 2, 63–77. Moore, R.A., Warren, R.L., Freeman, J.D., Gustavsen, J.A., Chénard, C., Friedman, J.M., Suttle, C.A., Zhao, Y., and Holt, R.A. (2011). The sensitivity of massively parallel
- 36. Microbes: An Unseen Majority Around Us | 21 sequencing for detecting candidate infectious agents associated with human tissue. PLoS One 6, e19838. Moran, M.A. (2009). Metatranscriptomics: eavesdrop- ping on complex microbial communities. Microbe 4, 329–335. Moreno-Paz, M., Gómez, M.J., Arcas, A., and Parro, V. (2010). Environmental transcriptome analysis reveals physiological differences between biofilm and planktonic modes of life of the iron oxidizing bacteria Leptospirillum spp. in their natural microbial commu- nity. BMC Genomics 11, 404. Motti, C. (2012). Environmental marine metabolomics: from whole organism system biology to ecosystem management. J. Mar. Sci. Res. Dev. 2, 3. Moyer, C.L., and Morita, R.Y. (2007). Psychrophiles and Psychrotrophs. In Encyclopedia Life Sciences (John Wiley Sons Ltd, Chichester, UK). Available from: http://www.els.net [doi: 10.1002/9780470015902. a0000402.pub2]. Muth, T., Benndorf, D., Reichl, U., Rapp, E., and Martens, L. (2012). Searching for a needle in a stack of needles: challenges in metaproteomics data analysis. Mol. Bio- syst. 9, 578–585. Naeem, S., and Li, S.B. (1997). Biodiversity enhances ecosystem reliability. Nature 390, 507–509. Nagalakshmi, U., Wang, Z., Waern, K., Shou, C., Raha, D., Gerstein, M., and Snyder, M. (2008). The transcrip- tional landscape of the yeast genome defined by RNA sequencing. Science 320, 1344–1349. Newman, D.J., and Cragg, G.M. (2007). Natural products as sources of new drugs over the last 25 years. J. Nat. Prod. 70, 461–477. Ohkuma, M. (2003). Termite symbiotic systems: efficient bio-recycling of lignocellulose. Appl. Microbiol. Bio- technol. 61, 1–9. Ortmann, A.C., and Suttle, C.A. (2009). Determination of virus abundance by epifluorescence microscopy. Methods Mol. Biol. 501, 87–95. Oshlack, A., and Wakefield, M.J. (2009). Transcript length bias in RNAseq data confounds systems biology. Biol. Direct. 4, 14. Pace, N.R. (1997). A molecular view of microbial diversity and the biosphere. Science 276, 734–740. Pan, S.Y., Pan, S., Yu, Z.L., Ma, D.L., Chen, S.B., Fong, W.F., Han, Y.F., and Ko. K.M.. (2010). New perspectives on innovative drug discovery: an overview. J. Pharm. Pharm. Sci. 13, 450–71. Parro, V., Moreno-Paz, M., and González-Toril, E. (2007). Analysis of environmental transcriptomes by DNA microarrays. Environ. Microbiol. 9, 453–64. Passalacqua, K.D., Varadarajan, A., Ondov, B.D., Okou, D.T., Zwick, M.E., and Bergman, N.H. (2009) Struc- ture and complexity of a bacterial transcriptome. J. Bacteriol. 191, 3203–3211. Petrić, I., Bru, D., Udiković-Kolić, N., Hršak, D., Philippot, L., and Martin-Laurent F. (2011). Evidence for shifts in the structure and abundance of the microbial com- munity in a long-term PCB-contaminated soil under bioremediation. J. Hazard Mater. 195, 254–60. Petti, C.A. (2007). Detection and identification of micro- organisms by gene amplification and sequencing. Clin. Infect. Dis. 44, 1108–1114. Pinto, A.C., Melo-Barbosa, H.P., Miyoshi, A., Silva, A., and Azevedo, V. (2011). Application of RNA-seq to reveal the transcript profile in bacteria. Genet. Mol. Res. 10, 1707–18. Poretsky, R.S., Hewson, I., Sun, S., Allen, A.E., Zehr, J.P., and Moran, M.A. (2009). Comparative day/night metatranscriptomic analysis of microbial communities in the North Pacific Subtropical Gyre. Environ. Micro- biol. 11, 1358–1375. Porter, K.G., and Feig, Y.S. (1980). The use of DAPI for identifying and counting aquatic microflora. Limnol. Oceanogr. 25, 943–948. Prakash, T., and Taylor, T.D. (2012). Functional assignment of metagenomic data: challenges and applications. Brief. Bioinf. 13, 711–27. Psenner, R. (1990). From image analysis to chemical anal- ysis of bacteria: a long-term study? Limnol. Oceanogr. 35, 234–237. Ram, R.J., Verberkmoes, N.C., Thelen, M.P., Tyson, G.W., Baker, B.J., Blake II, R.C., Shah, M., Hettich, R.L., and Banfield, J.F. (2005). Community prot- eomics of a natural microbial biofilm. Science 308, 1915–1920. Rappe, M.S., and Giovannoni, S.J. (2003). The uncul- tured microbial majority. Annu. Rev. Microbiol. 57, 369–394. Rasmussen, S., Nielsen, H.B., and Jarmer, H. (2009). The transcriptionally active regions in the genome of Bacil- lus subtilis. Mol. Microbiol. 73, 1043–1057. Raz, T., Kapranov, P., Lipson, D., Letovsky, S., Milos, P.M., and Thompson, J.F. (2011). Protocol dependence of sequencing-based gene expression measurements. PLoS One 6, e19287. Roh, S.W., Abell, G.C., Kim, K.H., Nam, Y.D., and Bae, J.W. (2010). Comparing microarrays and next-gener- ation sequencing technologies for microbial ecology research. Trends Biotechnol. 28, 291–299. Roselló-Mora, R., and Amann, R. (2001). The species concept for prokaryotes. FEMS Microbiol. Rev. 25, 39–67. Rothberg, J.M., and Leamon, J.H. (2008). The develop- ment and impact of 454 sequencing. Nat. Biotechnol. 26, 1117–1124. Rothschild, L.J., and Mancinelli, R.L. (2001). Life in extreme environments. Nature 409, 1092–1101. Sanger, F., Nicklen, S., and Coulson, A.R. (1977). DNA sequencing with chain-terminating inhibitors, Proc. Natl. Acad. Sci. U.S.A. 74, 5463–5467. Schleifer, K.H. (2009). Classification of Bacteria and Archaea: past, present and future. Syst. Appl. Micro- biol. 32, 533–542. Schmid, A., Dordick, J.S., Hauer, B., Kiener, A., Wubbolts, M., and Witholt, B. (2001). Industrial biocatalysis today and tomorrow. Nature 409, 258–268. Schmidt, T.M., DeLong, E.F., and Pace, N.R. (1991). Analysis of a marine picoplankton community by 16S rRNA gene cloning and sequencing. J. Bacteriol. 173, 4371–4378. Schneider, M.V., and Orchard, S. (2011). Omics technolo- gies, data and bioinformatics principles. Methods Mol. Biol. 719, 3–30.
- 37. Jiwaji et al. 22 | Schopf, J.W. (2001) Cradle of Life – The Discovery of Earth’s Earliest Fossils. (New Jersey, USA: Princeton University Press). Selinger, D.W.,Cheung, K.J., Mei, R., Johansson, E.M., Richmond, C.S., Blattner, F.R., Lockhart, D.J., and Church, G.M. (2000). RNA expression analysis using a 30 base pair resolution Escherichia coli genome array. Nat. Biotech. 18, 1262–1268. Seshardi, R., Kravitz, S.A., Smarr, L., Gilna, P., and Frazier, M. (2007). CAMERA: A Community Resource for Metagenomics. PLoS Biol. 5, e75. Shanmughapriya, S., Kiran, G.S., Selvin, J., Thomas, T.A., and Rani, C. (2009). Optimization, purifica- tion,andcharacterizationofextracellularmesophilic alkaline cellulase from sponge-associated Marino- bacter sp. MSI032. Appl. Biochem. Biotechnol. 162, 625–640. Sharma, C.M., Hoffmann, S., Darfeuille, F., Reignier, J., Findeiss, S., Sittka, A., Chabas, S., Reiche, K., Hack- ermuller, J., Reinhardt, R., Stadler, P.F., and Vogel, J. (2010). The primary transcriptome of the major human pathogen Helicobacter pylori. Nature 464, 250–255. Shendure, J., and Ji, H. (2008). Next-generation DNA sequencing. Nat. Biotechnol. 26, 1135–1145. Siggins, A., Gunnigle, E., and Abram, F. (2011). Explor- ing mixed microbial community functioning: recent advances in metaproteomics. FEMS Microbiol. Ecol. 80, 265–280. Simon, M., and Azam, F. (1989). Protein content and protein synthesis rates of planktonic marine bacteria. Mar. Ecol. Progr. Ser. 51, 201–213. Sogin, M.L., Morrison, H.G., Huber, J.A., Welch, D.M., Huse, S.M., Neal, P.R., Arrieta, J.M., and Herndl, G.J. (2006). Microbial diversity in the deep sea and the underexplored ‘rare biosphere’. Proc. Natl. Acad. Sci. U.S.A. 103, 12115–12120. Sorek, R., and Cossart, P. (2010). Prokaryotic transcrip- tomics: a new view on regulation, physiology and pathogenicity. Nat. Rev. Genet. 11, 9–16. Stewart, E.J. (2012). Growing unculturable bacteria. J. Bacteriol. 194, 4151–60. Streit, W.R., Daniel, R., and Jaeger, K.E. (2004). Pros- pecting for biocatalysts and drugs in the genomes of non-cultured microorganisms. Curr. Opin. Biotech- nol. 15, 285–290. Stres, B. (2007). The relationship between total and cul- turable bacteria in cold soils. Acta agriculturae Slov. 90, 25–31. Struvay, C., and Feller, G. (2012). Optimization to Low Temperature Activity in Psychrophilic Enzymes. Int. J. Mol. Sci. 13, 11643–11665. Suenaga, H., Ohnuki, T., and Miyazaki, K. (2007). Func- tional screening of a metagenomic library for genes involved in microbial degradation of aromatic com- pounds. Environ. Microbiol. 9, 2289–2297. Synowiecki, J. (2010). Some applications of thermophiles and their enzymes for protein processing. Afr. J. Bio- technol. 9, 7020–7025. Thomas, R.K., Nickerson, E., Simons, J.F., Jänne, P.A., Tengs, T., Yuza, Y., Garraway, L.A., LaFramboise, T., Lee, J.C., Shah, K., et al. (2006). Sensitive mutation detection in heterogeneous cancer specimens by mas- sively parallel picoliter reactor sequencing. Nat. Med. 12, 852–855. Thomson, B., Ostle, N., McNamara, N., Bailey, M., White- ley, A., and Griffiths, R. (2010). Vegetation affects the relative abundances of dominant soil bacterial taxa and soil respiration rates in an upland grassland soil. Microb. Ecol. 59, 335–343. Toledo-Arana, A., Dussurget, O., Nikitas, G., Sesto, N., Guet-Revillet, H., Balestrino, D., Loh, E., Gripenland, J., Tiensuu, T., Vaitkevicius, K., et al. (2009). The Lis- teria transcriptional landscape from saprophytism to virulence. Nature 459, 950–956. Torres, T.T., Metta, M., Ottenwalder, B., and Schlotterer, C. (2008). Gene expression profiling by massively parallel sequencing, Genome Res. 18, 172–177. Tuffin, M., Anderson, D., Heath, C., and Cowan, D.A. (2009). Metagenomic gene discovery: how far have we moved into novel sequence space? Biotechnol. J. 4, 1671–83. Uchiyama, T., and Watanabe, K. (2007). The SIGEX scheme: high throughput screening of environmental metagenomes for the isolation of novel catabolic genes. Biotechnol. Genet. Eng. Rev. 24, 107–116. Uchiyama, T., Abe, T., Ikemura, T., and Watanabe, K. (2005). Substrate-induced gene-expression screening of environmental metagenome libraries for isolation of catabolic genes. Nat. Biotechnol. 23, 88–93. Ulukanli, Z., and Digrak, M. (2002). Alkaliphilic Micro- organisms and Habitats. Turk. J. Biol. 26, 181–191. Umrania, V.V. (2006). Bioremediation of toxic heavy metals using acidothermophilic autotrophies. Biore- sour. Technol. 97, 1237–1242. Venter, J.C., Remington, K., Heidelberg, J.F., Halpern, A.L., Rusch, D., Eisen, J.A., Wu, D., Paulsen, I., Nelson, K.E., Nelson, W., et al. (2004). Environmental genome shotgun sequencing of the Sargasso Sea. Science 304, 66–74. Vieites, J.M., Guazzaroni, M.E., Beloqui, A., Golyshin, P.N.,andFerrer,M.(2009).Metagenomicsapproaches in systems microbiology FEMS Microbiol. Rev. 33, 236–255. van Vliet, A.H. (2010). Next generation sequencing of microbial transcriptomes: challenges and opportuni- ties. FEMS Microbiol. Lett. 302, 1–7. Wang, H.B., Zhang, Z.X., Li, H., He, H.B., Fang, C.X., Zhang, A.J., Li, Q.S., Chen, R.S., Guo, X.K., Lin, H.F., et al. (2011). Characterization of metaprot- eomics in crop rhizospheric soil. J. Proteome Res. 10, 932–940. Wang, Z., Gerstein, M., and Snyder, M. (2009). RNA-Seq: a revolutionary tool for transcriptomics. Nat. Rev. Genet. 10, 57–63. Wanunu, M. (2012). Nanopores: A journey towards DNA sequencing. Phys. Life Rev. 9, 125–58. Wasmund, K., Burns, K.A., Kurtboke, D.I., and Bourne, D.G. (2009). Novel alkane hydroxylase gene (alkB) diversity in sediments associated with hydrocarbon seeps in the Timor Sea, Australia. Appl. Environ. Microbiol. 75, 7391–7398. Wilson, E.O. (1997). Introduction. in biodiver- sity. In Understanding and Protecting our
- 38. Microbes: An Unseen Majority Around Us | 23 Biological Resources, Reaka-Kudla, M.L., Wilson, D.E., and Wilson, E.O., eds. (Washington DC, USA: Joseph Henry Press) pp. 1–3. Wintzingerode, F., Gobel, U.B., and Stachebrandt, E. (1997). Determination of microbial diversity in environmental samples: pitfalls of PCR based rRNA analysis. FEMS Microbiol. Rev. 21, 213–229. Woese, C.R. (1987). Bacterial evolution. Microbiol. Rev. 51, 221–271. Woese, C.R., Kandler, O., and Wheelis, M.L. (1990). Towards a natural system of organisms: Proposal for the domains Archaea, Bacteria, and Eucarya. Proc. Natl. Acad. Sci. U.S.A. 87, 4576–4579. Wooley, J.C., and Ye, Y. (2009). Metagenomics: Facts and Artifacts, and Computational Challenges. J. Comput. Sci. Technol. 25, 71–81. Wurtzel, O., Sapra, R., Chen, F., Zhu, Y., Simmons, B.A., and Sorek, R. (2009). A single-base resolution map of an archaeal transcriptome Genome Res. 20, 133–141. Xu, C., Liu, L., Zhang, Z., Jin, D., Qiu, J., and Chen, M. (2013) Genome-scale metabolic model in guiding metabolic engineering of microbial improvement. Appl. Microbiol. Biotechnol. 97, 519–539. Xu, M., Xiao, X., and Wang, F. (2008). Isolation and characterization of alkane hydroxylases from a metagenomic library of Pacific deep-sea sediment. Extremophiles 12, 255–262. Xu, M., Wu, W.M., Wu, L., He, Z., Van Nostrand, J.D., Deng, Y., Luo, J., Carley, J., Ginder-Vogel, M., Gentry, T.J., et al. (2010) Responses of microbial community functional structures to pilot-scale uranium in situ bioremediation. ISME J. 4, 1060–1070. Yeh, W.-K., Yang, H.-C., and McCarthy, J.R. (2011). Enzyme Technologies: Metagenomics, Evolution, Biocatalysis, and Biosynthesis (New Jersey, USA: John Wiley Sons, Inc.) Yergeau, E., Kang, S., He, Z., Zhou, J., and Kowalchuk, G.A. (2007). Functional microarray analysis of nitro- gen and carbon cycling genes across an Antarctic latitudinal transect. ISME J. 1, 163–179. Yergeau, E., Bokhorst, S., Kang, S., Zhou, J., Greer, C.W., Aerts, R., and Kowalchuk, G.A. (2012). Shifts in soil microorganisms in response to warming are consistent across a range of Antarctic environments. ISME J. 6, 692–702. Yoder-Himes, D.R., Chain, P.S., Zhu, Y.,Wurtzel, O., Rubin, E.M., Tiedje, J.M., and Sorek, R. (2009). Map- ping the Burkholderia cenocepacia niche response via high-throughput sequencing. Proc. Natl. Acad. Sci. U.S.A. 106, 3976–3981. Zhu, L., Zhu, Y., Zhang, Y., and Li, Y. (2012). Engineering the robustness of industrial microbes through syn- thetic biology. Trends Microbiol. 20, 94–101. Zillig, W. (1991). Comparative biochemistry of Archaea and Bacteria. Curr. Opin. Genet. Dev. 1, 544–551.
- 40. 2 Prokaryotic Genome Sequencing and Assembly Morag Graham, Gary Van Domselaar and Paul Stothard Abstract Researchers can now readily obtain millions of sequencereadsfromthegenomesoftheirfavourite prokaryotic organisms thanks to the development of next-generation sequencing technologies. Through sequence assembly, it is possible to reconstruct large portions of a genome from the overlapping sequence reads. However, assembly is challenging because the sequence reads are generally quite short and genomes often contain internally repeated segments that may confound the complete reconstruction of a genome from its constituent reads. There are different approaches for addressing these challenges that involve, for example, more advanced assembly tools, refer- ence genome sequences, and directed follow-up sequencing. Regardless of the strategy employed there are many steps and programs involved, and the final outputs need to be annotated and inter- preted with the known shortcomings of the data and methodologies in mind. Introduction Modern sequence technologies – termed ‘next generation’ or ‘next-gen’ sequencing (NGS) – have revolutionized the field of biology with their ability to rapidly and cheaply generate vast amounts of genomic sequence data. Up until the last half-decade, generating whole-genomic sequencedatahasrequiredsignificantinvestments of time and resources generally available only at large sequencing centres; today NGS allows even small laboratories to routinely generate genomic sequence data for organisms under study. The widespread adoption of these technologies has created tremendous new opportunities for bio- logical research, but also new challenges owing to the bioinformatics required to process the vast amounts of sequence data that the platforms so readily generate. Current NGS technologies generate read lengths that sample merely a small fraction of the genome size of most microorganisms. Although it is possible to identify and annotate coding sequence regions and other features of interest on individual reads (as is often done in the context of metagenomics research), there is much to be gained from combining overlapping reads into larger contiguous sequences, or contigs, through a process called sequence assembly. The creation of contigs allows for more accurate and complete annotation of genomic features, and also permits more in-depth analyses of sequence evolution, gene structure, and various other sequence prop- erties and features (see Chapter 4 for further discussion). Initial sequence assembly approaches used a labour intensive technique called primer walking to iteratively extend and connect contigs. Later, a higher throughput, automated approach called shotgun sequence assembly was introduced. Today, most genome sequencing projects primar- ily adopt the shotgun sequencing strategy or a hybrid strategy utilizing both shotgun sequencing and primer walking to acquire genomes. Depend- ing on the goals of the sequencing project and the availability of a closely related reference genome sequence, so-called reference mapping may be employable, in which sequence reads from the newly sequenced genome are aligned and mapped to the reference genome to identify differences (and similarities) in the target genome.
- 41. Graham et al. 26 | In this chapter we provide an overview of sequencing technologies as well as the down- streamprocessingstepsthatcanbeusedtoconvert the raw reads into larger contigs suitable for anno- tation of genes and other genomic features (Fig. 2.1). Strategies for specific sequencing objectives are presented, which include suggestions for sequencing technology, library types, assembly methods and software. Given the rapid rate at which new sequencing technologies and bioinfor- matics tools are being produced, such strategies need to be regularly revised, but these guidelines should serve as a useful starting point. Lastly, the chapter is intended to introduce the method- ologies associated with genome sequencing and assembly; it not intended to be comprehensive with respect to existing bioinformatics tools. Sequencing technologies First generation sequencing approaches The first widely adopted approaches to DNA sequencing were developed in the mid-seventies. The pioneering method invented by Frederick Sanger and Alan Coulsen (Sanger and Coulson, 1975), commonly referred to as the Sanger method or chain-termination sequencing, is based on the incorporation of chain-terminating 2′,3′-dide- oxyribonucleotide 5′-triphosphates (also called dideoxynucleotides; ddNTPs) during replication of a single-stranded DNA template fragment. In this approach, the DNA fragment to be sequenced is separated into four different reactions containing radiolabelled DNA primer, DNA polymerase, and one of the four dideoxynucleotides (con- taining adenosine (ddATP), cytosine (ddCTP), guanosine (ddGTP), or thymidine (ddTTP), respectively). A dideoxynucleotide terminates the chain thereby blocking extension because other nucleotides cannot bind to it owing to its 3′-end modification. Thus, the polymerase extends the labelled DNA primer until a dide- oxynucleotide is randomly incorporated, at which point the sequencing reaction is terminated. These fragments are then denatured, separated by electrophoresis and visualized using radiography. The random incorporation of chain-terminating ddNTPs guarantees that every fragment length (correspondingtoeverypossiblesinglenucleotide addition) will be represented as a distinct band in the electropherogram, thus providing a mecha- nism to deduce the genomic sequence. A different approach, developed around the same time by Alan Maxam and Walter Gilbert (Maxam and Gil- bert, 1977) employs radiolabelling at the 5′ end of the DNA fragment to be sequenced, followed by chemical treatments to cleave the DNA fragment at specific residues or residue pairs. The fragments are separated by size using electrophoresis, and then detected by autoradiography. Analysis of . . . . . . . . . . Process Time Frame Microbial WGS Project Flow Figure 2.1 #5 /% * ]# #5 #5 * ^ # study design I * 7 # /% /% /% : ) 7 # D$ 7 % _ ! /%7 7 Sample collection ! !7 ) ` ! 7# . I # ! ! /% '% D$ /% culture 7 % 5 % % ! NGS library construction sequencing are described ] :00 7data analysis7 % j 7 % ) /% % % :0
- 42. Genome Sequencing and Assembly | 27 the fragment sizes permits reconstruction of the DNA fragment sequence. A large advancement was achieved in 1986 with the introduction of fluorescent labels to replace the radiolabelled DNA (Smith et al., 1986). In addition to being safer than radiolabelling, fluorescent labels facili- tate detection by optical systems that are more amenable to automation and miniaturization. The chemical cleavage approach developed by Maxam and Gilbert was initially more popular than the biochemical chain-terminator approach developed by Sanger and Coulson. However, improvements to the Sanger method, such as the introduction of dye-terminator sequencing where the ddNTPs are fluorescently labelled (instead of the DNA primer) allowing sequencing to occur in a single reaction, ultimately made the Sanger approach faster, easier to use, and better suited to automated sequencing. Today, the majority of first-generation automated sequence analysers use the dye-terminator sequencing approach. Automated sequencers employing Sanger sequencing chemistry, commonly called Sanger sequencers, apply capillary electrophoresis separa- tion and fluorescence detection to generate DNA sequences. Completed sequencing reactions undergo electrokinetic injection into the capil- lary sequencer, which separates the fluorescently labelled, dye-terminated sequences based on frag- ment size. The instrument records the fluorescent signals from each capillary, generating sequence traces, or chromatograms, that show the signal detected simultaneously for each of the four possible terminating bases at each position. On- board software performs the signal processing for nucleotide base calling and confidence scoring. High quality traces, typically 300–1200bases in length, are automatically converted into DNA sequence and reported as a string of nucleotides (a sequence read) along with their associated quality scores (often called Phred scores) (Ewing and Green, 1998; Ewing et al., 1998). A typical modern automated Sanger sequencer can accom- modate 96 or 384 DNA samples at a time and can run unattended with multiple sample plates queued, generating a theoretical throughput of nearly 1Mb per day at a cost of ~$2500 per Mb. Despite having been largely supplanted by next-generation sequencing (NGS), and being dramatically less cost-effective per base than NGS, the Sanger method is still the preferred approach for many small-scale projects, and remains in widespread use today. Next-generation sequencing (NGS) approaches The first genome to be sequenced was bacte- riophage PhiX174 (Sanger et al., 1977). This viral genome, with only 5375bp, is within the range of early (pre-automated) sequencing technologies. Prokaryotic genomes, with sizes ranging from 0.5–10Mb, require higher throughput automated methods. Advances in sequence analysis made duringtheearly1990sincreasedthethroughputof automated Sanger sequence analysers sufficiently to make practical the generation of whole-genome sequences of prokaryotes, and even higher organ- isms, including the first 3.3Gb draft human genome, completed in 2000 (Lander et al., 2001; Venter et al., 2001). However, these early efforts required massive resources to complete, in some cases requiring hundreds of scientists, thousands of Sanger sequencing instruments, billions of dollars, and many years of labour. The continuous demand for low-cost, low-labour, whole-genome sequence data combined with the limitations of automated Sanger sequencing drove the devel- opment of newer next-generation sequencing (NGS) platforms, of which a variety are now commercially available. In these, nanofluidic mechanisms provide small-volume reactions for economical molecular sequencing at a massively parallel scale. NGS occurs via in situ sequencing by synthesis (SBS). Most platforms apply a DNA polymerase or ligase enzymes to simultaneously synthesize the new DNA strands. SBS most often occurs on multiple identical copies of a DNA template molecule, usually amplified as a group on beads (or an isolated surface). Alternatively, SBS can occur as a single molecule-based process in which sequencing detection occurs on a single molecule. Real-time SBS may occur, in which a free-running DNA polymerase is given all nucleotides required. More often, SBS platforms control the sequenc- ing process in a ‘stop-and-go’ iterative fashion – in order to assist with the identification of the incorporated nucleotide or oligonucleotide in
- 43. Graham et al. 28 | the growing strand. This can be accomplished by supplying substrates (nucleotides or short oligo- nucleotides) that are modified with an identifying tag (such as a fluorophore) and also reversibly blocked or else by providing only a single kind of substrate (e.g. dATP) at a time. Nucleotides and reagents are cyclically and individually washed across the immobilized templates, whereupon the specific incorporation of nucleotide(s) is detected by high-resolution digital imaging. Thereafter, reagents are washed away and a new reagent cycle is initiated. Thus, most NGS is currently achieved on spatially distinct, immobilized DNA template libraries via cycles of nucleotide incorporation/ washing/imaging. In contrast, single-molecule sequencing (SMS) requires no such nucleotide/ washing/imaging cycles; instead SMS directly monitors and detects nucleotide addition on a single DNA fragment strand. Other proprietary features of each NGS technology are related to the specific automated fluidic and optic technologies applied to capture each nucleotide series for each template library. Lastly, it is worth mentioning that regeneration of raw image data is less expen- sive than storage of raw data owing to the sheer size of NGS image data files produced; hence the collected raw data files are generally deleted once processed. All current commercial NGS technologies require DNA template preparation (NGS library construction), immobilization of said libraries, followed by massively parallel sequencing. The overall NGS procedure is summarized in Fig. 2.2. Proprietary nuances rest in how the stages are achieved. In general terms, bacterial genome sequencing is accomplished by highly redundant/ deep sequencing of random fragments making up the genome (Fig. 2.3). Random genomic library sequencing is sometimes referred to as ‘shotgun sequencing’, and is often achieved with ‘single-end’ sequencing of the library fragments (although not always – see below). In single-end sequenc- ing, the library fragment is sequenced exclusively from one direction. Random fragmentation of genomic DNA is achieved using mechanical shearing, sonication or enzymatic processes. The random DNA fragment ends are then made ame- nable to immobilization and/or sequencing via end polishing and incorporation of universal but platform-specific sequences (called ‘adaptors’). Following immobilization, each adaptor-contain- ing template library most often requires clonal amplification to achieve sufficient quantities prior to massively parallel sequencing and detection. Amplification can be performed on tethered libraries or in solution, but the efficiency of any NGS platform is tied to the relative assurance that each amplified library clone contains multiple copies of only a single and identical DNA frag- ment. The sole exception is the single-molecule sequencing approach commercialized as PacBio RS by Pacific Biosciences, which as the name sug- gests operates on a single DNA fragment. Modification of the standard ‘shotgun’ tem- plate library procedure can be applied to generate mate-pair libraries, comprised of originally dis- continuous sequence fragments from the genome separated by intervening nucleotides that are later enzymatically linked together in order to undergo sequencing. As such, mate-pair sequencing ™ TEMPLATE QC PREPARATION (randomly shear; end repair; size select) ™ NGS LIBRARY CONSTRUCTION (append sequencing adaptors + multiplex identifiers) ™ CLONAL AMPLIFICATION of Libraries (layout library on sequencing slide/wells) ™ MASSIVELY PARALLEL SEQUENCING (determine order identity of bases [at either end] of fragment) ™ RAW DATA ANALYSIS (image processing + base calling, with associated qualities) Figure 2.2 D) 5 /% !!#I 7 ) 5 * % . % I template QC and preparation7D ./% # I NGS library construction7 D/% ! /% = I clonal 7 % % %/% /% D = /%$$/% I massively parallel sequencing !% /% z raw data analysis7 /% % /% D$ # /% % %
- 44. Genome Sequencing and Assembly | 29 provides a cost-efficient means of co-sequencing discontinuous fragments (mates) from across the genome. Knowing that the mated reads must face each other (in most cases) across an average known gap size, even without determining the intervening sequence within the gap the mated reads can be applied to generate a ‘genomic scaf- fold’ that is useful for orienting gapped contigs (Fig. 2.3). Paired-end sequencing refers to the sequencing of both ends of DNA library frag- ments and is thus performed when mate-pair libraries are used. Paired-end sequencing can also be applied to fragments prepared through stand- ard, non-mate-pair (i.e. shotgun) template library procedures. In this case, the paired reads are from each end of the same NGS library fragment, and are thus separated on the original genome by only the small NGS fragment insert size. In the literature the term ‘paired-end’ is sometimes used to refer exclusively to this latter scenario; some- times for long-insert, mate-pair libraries (Roche pyrosequencing). Owing to differences in library Paired-end sequencing Genomic DNA Generate random genomic fragments Genomic scaffold Mate-pair reads Contigs Single-end sequencing “draft” genome “finished” genome Gap-filling Sequence polishing De novo assembly Mate pair library protocol NGS library construction Figure 2.3 z 7 7 = =%% D$ /% D ! D$ 7 =D{ | % .D$ $/% % single- endpaired-end/% # # {mate-pair| # 5 /% # 7 5 %% # ! % . # /% ) . # 5 % %| % 5 5 :;; de novo assembled { |% { | ^7 % 7% % * 7# in silico . Y) ! % # ! (' = # % 5 D # . `% ! ] . /% % ' ;7:;$% % /% % _/% 7# !=! { |' %%= % ) 7%
- 45. Graham et al. 30 | preparation protocols and sequencing method- ologies, the expected orientations of the reads with respect to the underlying genome can differ. It is important to be aware of these methodologi- cal and technology-specific details when working with the sequence reads. It is also important to note that discovered inconsistencies between mate-pair read orientations and a given reference sequence may indicate structural variation in the new sequence or the presence of erroneous chimeric pairings inadvertently generated during library preparation (see ‘Evaluating a genome assembly’ and Chapter 3 for more detail). Failure to incorporate a single nucleotide in a given NGS cycle may result in off-phasing, in which some library molecules lag in their exten- sion. When too many nucleotides are added, it is called pre-phasing. Pre-phasing and off-phasing cause the extracted signal intensities for a given cycle to be muddled with noise from the preced- ing or following cycles of sequencing, respectively. Pre-phasing and off-phasing contribute to loss of synchrony in the readout of the sequence copies for a given library fragment. The detected signals associated with the growing DNA strand become increasingly muddled over time, resulting in a quality drop-off for the called bases. Sequence base calling accuracy generally declines as the NGS read length increases; eventually no bases can be accurately called. As phasing susceptibil- ity varies with each NGS technology, the various NGS platforms vary in the lengths and qualities of their output sequence reads, and whether all reads are of the same or of variable lengths. Owing to this quality variation, it is important and neces- sary to assess data quality at multiple instances within any NGS project (Fig. 2.4). With the caveat that this chapter merely represents a snapshot in time – as dramatic advancements in sequencing technology are con- tinuously occurring – here we review, in general terms, the main commercialized NGS technol- ogy platforms in use today. Table 2.1 contains a summary of the performance metrics (current at Primary Analysis Analysis of hardware-generated data, sequencing run statistics, etc. Generate sequence reads quality scores Secondary Analysis QA QC filtering of raw reads (read cleaning) Alignment + mapping or de novo assembly of reads QA variant calling on aligned + mapped reads Tertiary Analysis Multi-sample processing QA/QC of variant calls; mate-pair/assembly assessment Genome annotation; feature identification Annotation filtering of variants Data aggregation curation; data improvement Interpretation / hypothesis generation The “sense making” Figure 2.4 !! ) 5 /% * I primary analysis7 # 5 7% /% % I secondary analysis7 /% /% % % /% /% % = % /% # !/% %de novo I tertiary analysis7# % 7 % 5 7 % = 7 /% = ! %
- 46. Genome Sequencing and Assembly | 31 Table 2.1 ' ) 5 /% Readers are !# z% et al., 0G-0j z et al., 0G-0j }% et al., 0G-0 % % latest = as = !! B are % # ! #5 /% assembly a 5 $ 5% 5 /% $ % /% (/% $/% by $/% by ) I % ( RS IIc ( = I ( z GS :z~ 3_ B$/ 0GGG, $/ I% I $,z 33GG€ I ;H;G~z Read 3GGG ! % 0G,GGG ;3`GG H3G ! % -GGG 0-GG B$/, 003G $/ H3 03G mate GG`FGG % FFFFF‚ }3G # CSS F9‚ }0G FFF‚ }0G F9‚ };G FFF‚ ƒ};G, FFFF‚ # ECC % FFFFF‚ }3G Reads % a ;3,GGG reads, % H3,GGG -0-G6 --G6 @GGG-G6 B$/, -H-G6 $/ GG-G6 F@ ;9 -@ % be /%% % Time % b ;G`-0G 0 0 11days B$/, 1day $/ H`-0 ; ' -5 bases $„ „0 „- „-G „GG3 „G-3 „G-; „0GG ! D = : Y are z $ % % ! D base = z# Small : D 5 ! z reads B %% B % z# I :#' % B /% , 6 D$ = ! I % size D mated z# yield at % B Y) ! /% B Y) ! B$/ Hybrid assemblies Y) ! $ Hybrid assemblies Y) ! /% I /% * a Reads % # =% % b % times are data c ( = % ( RS II # be ! the time of writing) for the various sequencing approaches described in this section. Readers are also referred to recent reviews (Liu et al., 2012; Loman et al., 2012; Quail et al., 2012) and the NGS technology manufacturers for the latest specifications. Pyrosequencing PyrosequencingdiffersfromSangersequencingin that the former detects direct release of a pyroph- osphate molecule as a by-product of nucleotide incorporation; whereas the latter detects a fluores- cent signal emitted from size-sorted fluorescently
- 47. Graham et al. 32 | labelled sequence fragments. The pyrosequencing technique, first developed in 1996 (Ronaghi et al., 1996), has been modified for miniaturization and automation and was incorporated into the Roche/454 line of Genome Sequencers (GS) madecommercialin2005(Marguliesetal.,2005). The commercialized platforms (GS FLX/GS Jr) are based on the sequencing by synthesis princi- ple. As previously described, the first step in the procedure is referred to as template preparation. Template preparation begins with the random shearing of whole genomic DNA into random, shorter fragments, typically by mechanical shear- ing, nebulization or by enzymatic processes. Adaptors containing sequences for (A) amplifica- tion and (B) pyrosequencing are then ligated to the ends of the fragment. Fragments containing the A/B adaptors are recovered and captured onto oligo-adapted capture beads under conditions that favour the binding of one DNA fragment molecule per bead. The bead-fragment library and PCR ampli- fication reagents are placed into a ‘water-in-oil’ emulsion under conditions that favour one frag- ment-bound bead per water droplet microreactor. Emulsion PCR (emPCR) is employed to generate millions of bead-bound, clonally amplified single stranded genomic templates in each microreactor. The emulsion is then broken and the bead-bound templates recovered. After immobilization and enrichment, the next GS pyrosequencing procedural step is the actual sequencing of bead- bound single stranded DNA templates. DNA beads are incubated with a mixture containing DNA polymerase and deposited into a PicoTiter Plate™ (PTP) containing millions of wells, with each well accommodating a single sequencing bead. Enzyme beads containing sulfurylase and luciferase are then added to the wells, along with packing beads. On instrument, the four DNA deoxyribonucleotide triphosphates (dNTPs) are added sequentially in a controlled flow order (e.g. T then C, A, G) across the PTP plate, during which only those dNTPs complementary to the template strand are incorporated into the growing strand. Incorporation of one or more nucleotides liberates hydrogen and pyrophosphate. The gen- erated pyrophosphate is converted into a light signal by the enzymatic action of sulfurylase and luciferase, and the light signal is recorded by a charge-coupleddevice(CCD)cameraattheendof every nucleotide flow. The total light signal gener- ated is proportional to the number of nucleotides incorporated into the growing strand during the single flow cycle; thus, homonucleotide stretches generate an enhanced signal relative to single nucleotide incorporations, and corresponding to the number of added nucleotides. The successive addition of dNTPs creates a series of light signals for each well, which at the completion of each flow are recorded initially as TIFF images with signal quality information. When non-complementary dNTPs are flowed and are not incorporated, no light signal is generated for that nucleotide flow and the well has no light emitted for that flow cycle. Later, these TIFF images are computation- ally processed yielding called bases in a sequence ‘flowgram’ with corresponding quality scores for each called base. A major disadvantage for pyrosequencing is difficulty in accurately quantitating signal strengths for stretches of homonucleotides owing to a phenomenon of light signal decay; homo- nucleotide tracts (termed homopolymers) are accuratelycalledonlytoroughlyeightconsecutive nucleotides. When inaccurate base calling occurs, the result is an incorrectly inferred insertion or deletion (indel) error. Semiconductor sequencing Semiconductor sequencing uses chemistry simi- lar to pyrosequencing to generate sequence reads. The major difference between the two approaches is that semiconductor sequencing employs instru- mentation that detects the produced hydrogen ions during nucleotide incorporation, rather than pyrophosphate. The resultant concentration of hydrogen ion liberated during dNTP incorpora- tionaffectsthepHwithinthemicrowell,whichisin turn detected by an ion-sensitive field-effect tran- sistor. As with pyrosequencing, the concentration of hydrogen ions is proportional to the number of incorporated dNTPs in the growing strand, and as such the semiconductor sequencing platform is stillsubjecttohomopolymer-inducedindelerrors. However, as the nucleotide incorporation signal is measured directly as an electrical pulse, no signal conversion is required. The resulting benefit is
- 48. Genome Sequencing and Assembly | 33 substantially reduced run times relative to Roche pyrosequencing and simplified detector construc- tion. Although work is under way to enhance clonal enrichment, semiconductor sequencing currently still requires emPCR for library enrich- ment prior to sequencing. The technology has been incorporated into the Life Technologies line of commercial sequencers, including the Ion Torrent Personal Genome Machine® (PGM™) and Ion Proton™. Sequencing by synthesis As with Roche pyrosequencing and semiconduc- tor sequencing, Illumina sequencing is based on theinsitusequencingbysynthesis(SBS)principle. The main distinct features of the Illumina-based SBS approach are a solid-support amplification method deployed for template fragment library enrichment(termed‘bridge-PCR’)anduseoffluo- rophore-labelled reversible terminator nucleotide chemistryduringmassivelyparallelSBS.Template preparation incorporates universal adaptors that are complementary to ‘anchor’ oligonucleotides covalently linked to an immobilized glass surface termed a ‘flow cell’. Adapted genomic fragment libraries are then affixed to the glass slide surface via annealing to these anchor oligos on the flow cell. After annealing, adaptor-containing template DNA molecules are then clonally amplified in a modified, isothermal PCR reaction (the bridge- PCR step) in which the DNA molecules are free to flex and form a ‘bridge’ with a second nearby flow cell functionalized oligonucleotide. Formed bridges then undergo isothermal extension in situ ontheglasssurface.This‘clustergeneration’process generates individualized clusters each containing approximately 1000 copies of identical, clonally amplified DNA molecules right on the flow cell surface, and millions of library clusters with a diameter of ~1μm per flow cell. DNA clusters are then readied for sequencing via denaturation to a single molecular strand (termed ‘linearization’) followed by blocking of the free 3′ ends of the cluster fragments and hybridization of a sequenc- ing primer. During each sequencing cycle, all four (reversibly) 3′-blocked dNTPs with corre- sponding fluorescent dye-bound terminators are simultaneously flowed and incorporated into the growing nucleotide chain by DNA polymerase at each cluster. The dye-terminator dNTPs ensure that each chain is extended by a single nucleotide, which is identified via its (label) fluorescence emission. At the end of the flow cycle, the flow cell is digitally imaged by a CCD and then the fluores- cent dye is removed by washing, enabling another sequencing cycle. Base calling is achieved by measuring the signal intensity via fluorescence emission at each cluster during each sequencing cycle. Such base-by-base sequencing shows fewer base call- ing errors associated with homonucleotide runs than pyrosequencing. In contrast, this short read SBS technology is primarily subject to random base substitution errors and so-called G-motif (GGCxG) errors. Sequence reads generated with SBS technology are shorter than pyrosequencing reads, but are less expensive to generate and can be produced readily with higher throughput; fea- tures that have made this NGS platform extremely popular. Illumina Inc. markets a number of sequencers incorporating this technology; popu- lar models include the HiSeq line of instruments and the more recently introduced MiSeq Personal Sequencer. Sequencing by oligonucleotide ligation and detection The SOLiD (Sequencing by Oligonucleotide Ligation and Detection) system commercialized by Applied Biosystems applies a collection of fluorophore-labelled oligonucleotides and liga- tion enzyme chemistry to achieve sequencing. Sequencing is actually comprised of multiple rounds during which eight base probes contain- ing a fluorophore are sequentially ligated to the template sequence to build up a complementary strand. Each round consists of the following: a priming step; a repeated cycle of ligating probes to the template; excitation and imaging of the fluorescence emission; and lastly, cleavage of the fluorophore and a terminal part of the bound probe prior to the next cycle of chemistry. Probe/ fluorophore combinations have been designed to interrogate the first two of the eight ligated positions in the template, with each of four fluo- rophore colours used to indicate four of the 16 possible nucleotide pairs at these positions. The emission of the fluorophore for each template is
- 49. Graham et al. 34 | recorded and applied later to infer the nucleotide sequence of the template strand (output read). SOLiD template libraries are prepared via random fragmentation to an appropriate size range; end repair, followed by ligation of ‘P1’ and ‘P2’ DNA adaptors to the ends of the fragments. EmPCR is deployed to immobilize the adapted DNA onto ‘P1’-coated paramagnetic beads. High- density semi-ordered arrays (polonies) of the DNA templates are then generated by function- alizing the 3′-ends of libraries and immobilizing the beads onto a solid glass slide. Sequencing is achieved by cyclic ligation of the pool of uniquely labelled, partially degenerate, fluorescently labelled DNA octamers, containing all possible dinucleotide variations of the two-base ‘recogni- tion core’. When these complementary detection probes anneal to the immobilized template, they are ligated to the primer. After imaging the fluores- cence emission, strands not extended are capped and the fluorophore and last three bases are cleaved from the probe. With this cleavage, the strand is available for extension in the next cycle of ligation-based sequencing; but now being initi- ated 5 bases upstream (n minus 5; i.e. n−5) from the priming site. The repeated ligation of eight additional bases and the cleaving of a terminal three bases ensure that the pair of bases of tem- plate sequence being examined then moves on by five positions every cycle. After seven cycles, the complementary strand is melted from the template to leave the template ready to be primed for the next round. Each round applies a different primer such that positions interrogated on the target library by the probes change each time; for example, on the first round the first position of the probes on cycles 1, 2, … corresponds to the tem- plate sequence positions 1, 6, …; on round 2 these become template sequence positions 2, 7, …, etc. After seven sequencing cycles, the first sequenc- ing primer is removed and a second sequencing primer is hybridized to the template strand (at the n-1 site). In total, five distinct sequencing primers (n, n-1, n-2, n-3 and n-4) are required for SOLiD. The SOLiD system uses a complex algorithm to deconvolute the fluorescence signals from labelled oligonucleotides in the off-set sequencing series, and reports the output in ‘colour space’. This peculiar output data is one downside of SOLiD sinceitgenerallynecessitatestheuseofareference sequence and many bioinformatics programs do not support colour space reads. However, newer SOLiD instruments can generate conventional base-space output through the use of an ECC (Exact Call Chemistry) module. ECC augments the two-base-encoding chemistry (achieving 2× redundant sampling of each base by the dinu- cleotide recognition core structure of the octamer (detection) oligonucleotides) with an additional round of ligation, using an alternative set of three- base encoding probes interrogating three positions. Each cycle of this round then interrogates posi- tions 1, 2 and 4,… of each five-nucleotide block of the template. The same four fluorophores are used, each now indicating the presence of one of 16 of the possible 64 combinations of nucleotides at the positions interrogated. These three-base encoded colour calls are then used to detect and improve base miscalls made in previous (two-base encoded) rounds. After imaging, templates to which a probe failed to ligate have their previous probe de-capped (i.e. dephosphorylated), such that they cannot be extended in future cycles. Thus, owing to ECC, the SOLiD system achieves good sequencing accuracy and less ‘phasing’ issues compared to other NGS technologies. Several features of the SOLiD platform are noteworthy. The SOLiD chemistry is non-intui- tive. A simple and rapid on-instrument ‘Wildfire’ template preparation technology (released in 2012) has accelerated the previous 8hour sample preparation methodology to a 2hour preparation time,andhasreducedthecostperbaseofSOLiDa further 50%. Life Technologies markets the Wild- fire chemistry for use with 5500 W and 5500 XL W Genetic Analyzers. A downside of sequencing by ligation is a fundamental limitation in achiev- able read length, which restricts the technology to shorter reads relative to other conventional NGS technologies (75bp), which in turn complicates sequence assembly. 9' 9 ;0 sequencing Single-molecule sequencing (SMS) is considered the next evolution of sequencing. Although a number of SMS approaches are under develop- ment, to date only one has been commercialized.
- 50. Genome Sequencing and Assembly | 35 The single-molecule real-time (SMRT) sequenc- ing approach, developed by Pacific Biosciences, applieszero-modewaveguide(ZMW)technology to tightly constrain the signal area generated from the sequencing of a single DNA template within a specially fabricated, deep well nanostructure on a glass surface (Korlach et al., 2010). During library preparation, hairpin loop adaptors are ligated to both ends of double-stranded DNA inserts. A tethered Phi29 polymerase is a highly processive, strand-displacing enzyme capable of performing rolling circle amplification: this acts on topologi- cally circular SMRTbell™ DNA template libraries. Sequencing is performed on a chip containing SMRT cells, each containing 150,000 arrayed ZMWs. The four nucleotide bases are each con- structed with a unique fluorescent dye attached to the phosphate chain. Applying an anchored, unmodified polymerase that is physically tethered to the base of a deep ZMW well, only fluorescence intensity resulting from addition of a fluorescently labelled nucleotide at the bottom of the well is recorded. Upon base incorporation, the fluores- cent dye is cleaved. The dye then diffuses out of the small ZMV-detection volume such that signal is no longer detected. Nucleotide incorporation is recorded by CCD, in real-time, for each DNA template in the form of a time-trace of detected fluorescence intensity from each ZMW; the poly- merase then translocates to the next position and the sequencing cycle is repeated. The SMRT technology can generate long reads (average 5000 bases; up to 20,000 bases with the latest instrumentation, PacBio RS II). When first introduced, the platform exhibited relatively low read accuracy (~87%); however, this has since improved (Table 2.1). Continuous long reads (CLRs) are generated when the adaptors located within a raw sequence read from a ZMW are removed; the read is split into multiple ‘sub- reads’. A full-pass subread is a subread with two observed adjacent adaptors. Large DNA inserts allow for lower accuracy CLRs that can help span gaps or larger repetitive regions, and are gaining popularity for scaffolding existing de novo genome assemblies. For variant detection applications and to overcome high error rates associated with SMRT CLR sequencing, an alternative SMS sequencing approach termed ‘circular consensus sequenc- ing’ (CCS) may be applied. Again leveraging topologically circularized SMRTbell™ library constructs, for CCS – shorter inserts are used. A CCS sequence is generated by collapsing multiple subreads from a single ZMW to form a single, higher-accuracy consensus read; the same insert is read on both the sense and the antisense strands multiple times (redundancy varies with insert size), thereby improving the final consensus read accuracy. Using 1–2kb insert SMRTbell™ libraries and CCS, one can achieve error correction for the longer CLRs prior to assembly. The large footprint and capital expenditure requirement, as well as the earlier low read accuracy has limited the popularity of this SMS platform relative to other mainstream NGS technologies. However, SMRT sequencing errors generally occur randomly throughout the sequence read, which means high consensus accu- racy can be achieved at relatively low coverage relative to other NGS technologies. The platform offers short run times (30–180min) and samples need not be amplified prior to sequencing. Con- sequently, there is growing interest in the PacBio platform for rapid de novo microbial genome sequencing, structural variation analysis, and genome closure. Emerging sequencing technologies Novel sequencing technologies are constantly under development, and expected to offer faster and cheaper sequencing. Several additional approaches of single-molecule sequencing, through nanopore structures, have been demon- strated with the hope of rapidly sequencing larger (native) DNA fragments without the need for prior time-consuming library template work-ups. A commercial version of nanopore sequencing will likely emerge in the not too distant future. Owing to space constraints, the reader is referred to excellent summaries of emerging sequencing technologies provided elsewhere (Venkatesan and Bashir, 2011; Wanunu, 2012). Quality assurance of sequence reads Regardless of the sequencing technology applied to generate NGS data, before using the raw
- 51. Graham et al. 36 | sequences end users should conduct an objec- tive check to assess the quality of each dataset. For example, it is important to ensure that reads are free of adaptor sequences, contaminant sequences, notable artefacts such as low complex- ity or excessive duplicate reads, and sequences of unacceptably low quality; otherwise these issues mayconfounddownstreamanalyses.Thisnotonly ensures that the best quality data goes forward to later analysis, it may also save time in the long run by flagging problematic reads before downstream analysis problems are encountered. The adage ‘garbage in, equals garbage out’ is very true in the context of genome sequencing and assembly. A useful means for accomplishing assessment of NGS data soundness is FastQC (http://www. bioinformatics.babraham.ac.uk/projects/down- load.html#fastqc), a quality control tool for high throughput sequence data. FastQC takes an input file in FASTQ format (a widely used format for NGS data that includes read sequences and qual- ity scores) and runs a series of tests to generate a comprehensive collection of useful QC assess- ments, statistics, and plots for visualizing the data. Other NGS utilities are increasingly available, and now in pipeline format (Blankenberg et al., 2010; Dai et al., 2010; Brouwer et al., 2012). Users should standardly opt to process data by applying data analysis pipelines, as such systematic and recordable approaches makes documentation and troubleshooting of analysis procedures much easier. It is highly advisable to remove problematic datapriortoembarkingontime-consumingdown- stream analyses. Read quality is initially assessed by looking at the read length distribution, Phred quality distribution, nucleotide frequencies, and read complexity. Low-quality read sets (showing uncharacteristically short read distributions or low quality scores) should be investigated or pos- sibly regenerated prior to downstream analysis. Read cleaning is achieved either by read filtering, wherein reads that fail to meet certain criteria are removed, or by read trimming which removes low- quality 3′ bases. It is relatively easy to find adaptor contamination or biases produced during the library construction protocol by plotting the fre- quency of each nucleotide for each position or by conducting homology searches (e.g. via BLAST analysis). If all positions are not observed to exhibit equivalent nucleotide frequencies in what should be random genomic DNA fragments, then the presence of adaptors within the reads should be suspected and the reads may need trimming. By trimming and removing low quality reads, the average sequencing quality of the remaining reads is improved, albeit with the loss of some minimal information. Knowing that NGS technologies are subject to base calling inaccuracies, error correction to improve sequence accuracy can be attempted to correct errors in the individual reads. Programs with this aim collect all reads (or parts of reads) that correspond to the same (likely) genomic region based on multiple sequence alignments and assume that any low frequency and random changes observed have resulted from sequenc- ing errors. Error correction is made based on the majority nucleotide call in the read col- lection. Methods employing analysis of DNA k-mer (DNA ‘words’ of length k) frequencies in whole-genome sequence are also available. Read- ers are referred to an excellent review of error correction methods for NGS (Yang et al., 2013). These authors have made their evaluation tools available (http://aluru-sun.ece.iastate.edu/doku. php?id=ecr#reference). Another useful program for read cleaning is Nesoni (http://www.vicbioin- formatics.com/software.nesoni.shtml). Sequence assembly Here we provide a brief overview of sequence assembly and read mapping approaches. It is important to note that sequence assembly is cur- rently a very active area of bioinformatics research owingtotheavailabilityoflow-cost,high-through- put sequencing technologies. Consequently, there is a vast array of bioinformatics tools cur- rently available, and there will undoubtedly be better ones created as the field evolves. Many of these programs and their underlying algorithms are specialized for particular types of sequence technology or sequence libraries. Thus, it is important to choose programs and approaches that are appropriate for the dataset(s) in hand and questions of interest, and to also regularly revisit (and likely revise) assembly strategies as new
- 52. Genome Sequencing and Assembly | 37 software tools are constantly released. Moreover, inherent flexibility is encouraged as NGS data characteristics improve at a rapid pace with each technology enhancement (for example, increasing read lengths and accuracies), and as new technol- ogy becomes available. Hence, the field is likely to remain unsettled for some time. We discuss some of the currently available tools and their applications at the end of this chapter. Additional information is found in Chapter 3, ‘Genome assembly – methods, tools and improvement’. Primer walking Early genome sequence assembly was carried out using the technique of primer walking, first introduced in 1993 (Voss et al., 1993). The pro- cess begins with the preparation of a library of large genomic DNA fragments. Genomic DNA is partially digested by endonuclease restriction enzymes, the fragments separated by gel electro- phoresis, and the separated fragments ligated into large-insert DNA vectors such as cosmids, fos- mids, bacterial artificial chromosomes (BACs), or yeast artificial chromosomes (YACs). The vector DNAistheninsertedintoasuitablehostorganism – typically bacteria – and propagated for template augmentation and retrieval. Genomic libraries are designed to contain several-fold, random coverage of the genome. The large genomic fragments contained in the aforementioned, low-copy genomic clone librar- ies are not directly suitable for Sanger sequencing and generally require enrichment prior to sequencing. Template augmentation can be achieved by sub-cloning insert DNAs as random fragments into higher-copy sequencing vectors. DNA inserts from each low-copy vector are iso- lated and randomly sheared to a size of 2–10kb. These randomly overlapping fragments are then selectively subcloned into a sequencing vector library containing annealing sites complementary to the universal primers used to initiate sequenc- ing reactions. The sequencing vector libraries are then sequenced usually from both ends (on both strands) by Sanger sequencing to acquire the unknown sequence for each DNA insert. This initial sequencing step identifies the first ~1000 bases of genome sequence from each end. Primers are subsequently designed (based on the acquired novel insert sequence) to anneal to the newly acquired (known) end(s) of high-confidence sequence, and used to extend further into the unknown insert region in subsequent rounds of sequencing. The procedure is then repeated with each cycle of primer ‘walking’ into the unknown portion of the insert DNA until sufficient overlap is achieved and it is completely sequenced, or found to match sequence from another genomic library fragment with which it can be merged and extended. The main drawback of primer walking is the requirement for identifying and synthesiz- ing different primers after each new sequence data acquisition step, making it a slow and largely manual approach for acquiring unknown sequences and closing sequence gaps. Hence, to overcome this limiting primer synthesis/ sequence/hold/repeat strategy, shotgun sequenc- ing assembly was developed and has since become adopted as the routine, current-day strategy for genome sequence assembly. Shotgun sequence assembly The origins of whole-genome shotgun sequencing can be traced as far back as 1979 (Staden, 1979), although this initial method was suitable only for relatively small genomes (under 10kb). The first large-scale approach to whole-genome shotgun sequence assembly was used to assemble the 1.8Mb genome of the bacterium Haemophilus influenzae (Venter et al., 1996). Briefly, two librar- ies consisting of ~2kb and 15–20kb inserts were prepared from random, mechanically sheared DNA, and then nearly 30,000 dye-terminator sequencing reactions were completed over a period of three months. The more than 11Mb of accumulated sequence read data were assembled using the TIGR ASSEMBLER program, which uses a greedy merging heuristic (problem solving technique) to combine overlapping fragments. This software identifies and scores all possible overlaps between reads, with larger overlaps receiving higher scores. The two reads with the highest scoring overlap are then merged into a single sequence. This merging step is repeated using the previously merged sequences and reads until no more sequences can be combined. Approximately 210 H. influenzae contigs were
- 53. Graham et al. 38 | obtained in this manner, which were further merged into 140 contigs using lower-stringency sequence comparisons among contigs. The contigs were then ordered using a variety of confirmatory wet laboratory approaches such as Southern blotting and PCR, and the remaining gaps were filled with a primer walking strategy. This landmark achievement set the stage for the widespread adoption of shotgun sequencing, although present day sequencing and assembly methods have since matured. Modern de novo assembly approaches A challenge when reconstructing a genome sequence from overlapping sequence reads is that repeats cause ambiguities that make the task of deducing the correct assembly difficult (or even impossible) depending on the characteristics of the reads and of the repeats (Nagarajan and Pop, 2009). Although greedy assemblers (such as used for the H. influenzae genome) successfully apply heuristics to avoid some issues caused by repeat sequences, they have been superseded by more advanced programs that better consider the global relationships among reads, making them better suited to the abundant short reads generated by modern NGS instruments. Here we briefly describe two widely used assembly strategies: overlap–layout–consensus and De Bruijn graph. For more detailed descriptions of these and other assembly methods see Chapter 3, and recent reviews of assembly techniques for NGS data (Miller et al., 2010; Nagarajan and Pop, 2013). The overlap–layout–consensus (OLC) strategy takes advantage of techniques from the field of graph theory, which uses mathemati- cal structures, called graphs, to model pairwise relationships between objects. These graphs, unlike the plots used to visualize functions or relationships between variables, consist of nodes (also called vertices) some of which may be con- nected by links (also called edges) indicating a relationship between the nodes (overlaps in the case of sequence reads). Once data is depicted in this manner, a variety of algorithms can be used to analyse the graph and infer the relationships of the sequences represented by nodes. OLC assem- blers construct a graph by first comparing each read to every other read, to identify base overlaps using a ‘seed-and-extend’ approach. This step is computationally intensive, but lends itself to parallelization. Overlaps are then represented as links in the graph, connecting nodes (reads). The assembler then refines and interprets the graph to identify potential contigs, which are identified as nonintersecting paths in the graph. Finally, these paths are converted to consensus sequences (consisting of the most abundant series of nucleo- tides for each node at each overlapping position) through the alignment of the constituent reads. An unordered listing of non-intersecting contigs is generated, with gaps of unknown sequence (and of unknown size) between contigs. De Bruijn graph assemblers draw heavily from previous developments in the field of computing science, and also construct a graph, albeit from overlapping, fixed-size subregions of the reads called k-mers, where k can typically be specified by the user and is generally substantially shorter than the read lengths themselves. There is an important computational advantage to this approach, as all reads do not need to be explicitly compared to all others using sequence alignment tools but instead can be rapidly summarized using an efficient data structure called a hash table, which is in this case an organized set of sequences all of length k. The assembly process, as in the case of the OLC approach, involves refining and interpreting the graph and then returning sequences based on the remaining paths. De Bruijn graph assemblers look for exact overlaps of fixed length between fragments and are consequently sensitive to sequencing errors, although some sequence errors can be identified and eliminated by a topological analysisofthegraph.Thus,itisrecommendedthat reads be quality filtered and trimmed to remove low-confidence bases prior to assembly using a De Bruijn graph-based tool (Salzberg et al., 2012). WhileOLCandDeBruijnapproachesareboth theoretically capable of producing high-quality assemblies, the short sequence read lengths of NGS technologies make assembly much more challenging than for conventional (Sanger) sequence data. One of the reasons for the diver- sity of assembly programs available is that each applies different strategies to address particular real-world nuances associated with each NGS
- 54. Genome Sequencing and Assembly | 39 technology. For example, programs are adapted to the read lengths, error characteristics, or insert sizes produced by particular sequencing plat- forms. Just as the greedy algorithms employ extra strategies to deal with repeated elements, so too must these assembly approaches for NGS data; many assemblers are able to incorporate informa- tion from other sources, such as information from mate-pair reads. Newer iterative approaches, such as IMAGE (Tsai et al., 2010), PRICE (Ruby et al., 2013), using mate-pair (or paired-end) read information have been developed to facilitate contig gap closure. Other challenges include non-uniform sequence coverage, regions com- pletely lacking sequence data (non-sampled or of problematic quality), and intrinsic sequencing errors. Both the OLC and De Bruijn assemblers are able to recognize certain graph structures that tend to arise from such issues, and apply a variety of strategies to attempt to resolve problematic structures. Most assemblers have several adjust- able parameters that can sometimes drastically alter assembly outcome; these parameters need to be determined empirically for some bacterial datasets. For these reasons, extensive testing of assemblyprogramsandparametersonrealdata(as opposed to simulated data) is necessary to make informed decisions regarding their performance in bacterial genome assembly. Fortunately, there are ongoing efforts in the research community to standardize such parameter testing and improve genome assembly evaluation; for example, see below and Magoc et al. (2013). Evaluating a genome assembly Once a draft genome is assembled from sequence reads, it is important to examine the quality of the result prior to embarking on time consuming downstream work. An assembly represents a best approximation of the genome determined in silico, based on the input data and assembly parameters selected. Although there may be existing stud- ies of the performance of a particular assembly program, likely conducted through comparing high-quality finished genomes to those recon- structed automatically from raw or simulated sequence data, the results may not predict well the quality for another new genome assembly made for a different bacterial species harbouring different repetitive sequences. When such evalua- tions are available, owing to rapid evolution of the NGS field, characteristics of any newly acquired raw reads likely differ from those of data previ- ously used to evaluate the assembly software. Computational methods that reliably assess the accuracy of genome sequence assemblies are currently lacking, but becoming increasingly important as assembly processes become more automated and larger numbers of genomes are being simultaneously processed. A variety of assembly metrics have been used in the literature, and the choice of which metric(s) to evaluate and report is the subject of debate. Most genome assemblers report information on the numbers and sizes of the resultant contigs. Typically included among these measures is a value called the N50 size, which is the length of the smallest contig such that 50% of the total contig length is contained in contigs of N50, or greater. Larger N50 sizes are typically interpreted as indicating a better assembly; however, N50 sizes only tell part of the story. Incorrect merging of contigs will lead to larger N50 values but generate an assem- bly that is not consistent with the true genome sequence. Overall, such crude metrics are only qualitative and hence, insufficient to describe genome assembly completeness; as mentioned, they also overlook potential mis-assemblies. For this reason additional assembly metrics should be examined. If a reference genome is already availa- ble for the newly sequenced species, reads can be aligned and mapped to this reference. If no refer- ence genome is available, the reads can be aligned and mapped against the generated contigs – in order to examine characteristics such as aligned read depth, insert sizes between paired-end or mate-pair reads, and read pair orientation. The idea here is that mis-assemblies can be identified when inconsistent or unusual mapping results are noted, such as mate-pairs being separated within the in silico assembly by incorrect large distances or with incorrect pairs orientation. Fortunately, tools for conducting these types of genome assembly quality assessments are becoming available (Phillippy et al., 2008; Vezzi et al., 2012; Gurevich et al., 2013; Rahman and Pachter, 2013), which weigh the likelihood that an assembly is computed accurately by evaluating
- 55. Random documents with unrelated content Scribd suggests to you:
- 59. The Project Gutenberg eBook of Kaksintaistelu
- 60. This ebook is for the use of anyone anywhere in the United States and most other parts of the world at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this ebook or online at www.gutenberg.org. If you are not located in the United States, you will have to check the laws of the country where you are located before using this eBook. Title: Kaksintaistelu Author: Anton Pavlovich Chekhov Translator: Emil Mannstén Release date: February 17, 2013 [eBook #42116] Language: Finnish Credits: Produced by Jukka Aakula *** START OF THE PROJECT GUTENBERG EBOOK KAKSINTAISTELU ***
- 61. Produced by Jukka Aakula KAKSINTAISTELU Kirj. Anton Tshehov Venäjänkielestä [Duelj] suomentanut Emil Mannstén Arvi A. Karisto Oy, Hämeenlinna, 1921.
- 62. I. Oli kello kahdeksan aamulla — se aika, jolloin upseerit, virkamiehet ja matkustavaiset tavallisesti kuuman, tukahduttavan yön jälkeen uivat meressä ja sitten tulivat paviljonkiin juomaan kahvia tai teetä. Ivan Andreitsh Lajevski, laihahko, vaaleaverinen, kahdeksankolmatta ikäinen nuorimies, rahaministeriön virkamieslakini päässä ja tohvelit jalassa, tapasi uimaan mennessä rannalla kosolta tuttavia ja niiden joukossa ystävänsä, sotilaslääkäri Samoilenkon. Samoilenkolla oli iso pää, tukka hyvin lyhyeksi leikattu, kaulaa tuskin ollenkaan, punakat kasvot, tuuheat mustat kulmakarvat ja harmaa poskiparta; kun tähän vielä yhtyi turpea vartalo ja käheä soturin basso, niin eipä ihmettä, että hän vieraaseen, joka ensi kerran saapui paikkakunnalle, teki vastenmielisen vaikutuksen. Mutta kun ehti kulua pari kolme päivää ensimmäisestä tutustumisesta, niin alkoivat hänen kasvonsa näyttää erittäin herttaisilta, lempeiltä, jopa kauniiltakin. Kankeista liikkeistään ja karkeahkosta äänestään huolimatta hän oli ihmisenä sävyisä, tavattoman helläluontoinen, jalomielinen ja kohtelias. Kaikkia kaupunkilaisia hän sinutteli, kaikille hän antoi rahoja lainaksi, oli heidän lääkärinsä, toimitteli heitä naimisiin, sovitteli heidän pikku riitojaan, pani toimeen ulkoilmakekkereitä, joissa paistoi hiilillä lampaanlihamöykkyjä ja keitti
- 63. piikkieväkaloista mainiota kalalientä; aina hän puuhasi ja pyysi jonkun puolesta, ja aina hänellä oli jotakin iloitsemisen syytä. Yleisen mielipiteen mukaan hän oli moitteeton mies, jolla tiedettiin olevan vain kaksi heikkoutta: ensiksikin hän häpesi omaa hyväsydämisyyttään ja koetti peittää sitä tuimalla katsannolla ja tahallisella töykeydellä, ja toiseksi hän tahtoi, että välskärien ja sotamiesten tuli puhutella häntä ylhäisyydeksi, vaikka hänellä oli ainoastaan valtioneuvoksen arvo. — Vastaa minulle, Aleksander Daviditsh, yhteen kysymykseen, — virkkoi Lajevski, sittenkun he molemmat, hän ja Samoilenko, olivat astuneet olkapäitään myöten veteen. — Otaksutaan, että olet rakastunut naiseen ja asettunut hänen kanssaan elämään yhdessä; parin vuoden kuluttua sitten, kuten voi sattua, et enää häntä rakasta, vaan alat tuntea, että hän on sinulle vieras. Miten sinä sellaisessa tapauksessa menettelisit? — Perin yksinkertaisesti. Sanoisin: lähde, muoriseni, mille ilmansuunnalle halajat — ilman pitempiä puheita. — Helposti sanottu! Mutta jollei hänellä ole mihin mennä? Jos hän on yksinkertainen nainen, vailla sukulaisia, vailla rahaa, jos ei osaa tehdä työtä… — Siinä tapauksessa … anna hänelle viisisataa ruplaa käteen tahi maksa viisikolmatta ruplaa kuukaudessa — ja tyytyköön siihen. Onhan se selvää! — Otaksutaan, että sinulla on nuo viisisataa tahi kuukausittain viisikolmatta, mutta nainen, josta puhun, on älykäs, hieno, ylpeä luonnostaan. Tokko ottaisit tarjotaksesi hänelle rahoja? Ja missä muodossa?
- 64. Samoilenko aikoi jotakin vastata, matta samassa suuri aalto hulvahti heidän molempien ylitse, loiskahti rantaan ja vyöryi pauhaten takaisin pienten kivien yli. Ystävykset nousivat rannalle ja alkoivat pukeutua. — Tietysti, onhan hankalaa elää naisen kanssa, jota ei rakasta, — lausui Samoilenko, karistaen hiekkaa saappaastaan. — Mutta pohtikaamme asiaa ihmisyyden kannalta, Vanja. Jos asia koskisi minua, niin en ollenkaan näyttäisi, että olen lakannut rakastamasta, vaan eläisin hänen kanssaan hamaan kuolemaan asti. Hänen tuli äkkiä häpeä omia sanojaan; niiden vaikutusta lieventääkseen hän sitten virkkoi: — Minusta on sama, vaikka ei akkaväkeä olisi olemassakaan. Lempo heidät periköön! Ystävykset pukeutuivat ja menivät paviljonkiin. Täällä Samoilenko oli kuin kotonaan, ja hänellä oli siellä omat erikoiset astiansakin. Joka aamu hänelle tuotiin tarjottimella kuppi kahvia, korkeassa särmäisessä juomalasissa jäänsekaista vettä ja ryyppy konjakkia; hän joi ensin konjakin, sitten kuuman kahvin, sitten jäänsekaisen veden, ja se lienee maistunut aika hyvältä, sillä nämä juotuaan hän sai silmiinsä erikoisen kiillon, silitteli molemmin käsin poskipartaansa ja virkkoi merta pälyillen: — Suurenmoinen näky kerrassaan! Pitkän, hyödyttömissä ja ikävissä ajatuksissa vietetyn yön jälkeen, jotka eivät sallineet hänen nukkua, vaan ihan kuin lisäsivät yön kuumuutta ja pimeyttä, Lajevski tunsi itsensä masentuneeksi ja veltoksi. Kylpy ja kahvi eivät tehneet hänen oloaan paremmaksi.
- 65. — Jatketaanpa, Aleksander Daviditsh, taannoista keskusteluamme, — sanoi hän. — Minä en tahdo peitellä, vaan puhun sinulle suoraan, kuten ystävälle: välini Nadeshda Feodorovnan kanssa ovat huonot … tykkänään huonot! Suo anteeksi, että ilmaisen sinulle salaisuuteni, mutta minun täytyy saada keventää sydäntäni. Samoilenko, aavistaen mistä tulisi olemaan puhe, loi katseensa maahan ja alkoi sormillaan naputtaa pöytään. — Olen elänyt hänen kanssaan kaksi vuotta, ja nyt on rakkauteni häneen lopussa, — jatkoi Lajevski, — tai oikeammin, olen tullut käsittämään, etten koskaan ole rakastanutkaan… Nämä kaksi vuotta ovat olleet petosta. Lajevskilla oli tapana keskustellessaan tarkkaavasti katsella heleänpunaisia kämmeniään, pureskella kynsiään tahi likistää sormilla kalvosimiaan. Ja nytkin hän teki samoin. — Minä tiedän varsin hyvin, ettet sinä voi minua auttaa, mutta puhun sinulle sentähden, että meikäläisen kovaosaisen ja liikanaisen ihmisen ainoa pelastus on puhumisessa. Minun täytyy tehdä tiettäväksi jokainen tekoni, minun on löydettävä selitys ja puolustus järjettömälle elämälleni kenenkä hyvänsä teorioista, kirjallisista tyypeistä, siitä esimerkiksi, että me, aateliset, huononemme suvustamme, ja niin edelleen… Viime yönä esimerkiksi minä lohdutin itseäni koko ajan ajattelemalla: ah, kuinka oikeassa on Tolstoi, kuinka säälimättömän oikeassa! Ja minulle tuli siitä helpotusta. Todellakin suuri kirjailija! Sanottakoon mitä hyvänsä. Samoilenko, joka ei koskaan ollut lukenut Tolstoita, vaikka yhtämittaa oli ollut aikeissa lukea hänen teoksiaan, joutui hämille ja vastasi:
- 66. — Niin, kaikki kirjailijat kirjoittavat mielikuvituksesta, mutta hän ottaa kuvattavansa suoraan luonnosta… — Voi sentään, — huokasi Lajevski, — missä määrin sivistys onkaan meidät turmellut! Minä rakastuin naituun naiseen; hän samaten minuun… Alussa meillä oli suuteloita, hiljaisia iltoja, lupauksia, Spencer ja ihanteita ja yhteisiä harrastuksia… Mokoma vale! Todellisuudessa pakenimme hänen miestään, mutta valehtelimme itsellemme, että pakenimme sivistyneen elämämme tyhjyyttä. Tulevaisuus kangasti meille seuraavanlaatuisena: aluksi Kaukaasiassa, kunnes perehdymme maahan ja ihmisiin, minä otan virastotakin ylleni ja alan hoitaa virkaa, sitten otamme kappaleen perkaamatonta maata, rupeamme tekemään työtä otsamme hiessä, istutamme viinitarhan, muokkaamme pellon, ynnä muuta. Jos minun sijassani olisit ollut sinä tai tuo eläintieteilijä von Coren, olisitte Nadeshda Feodorovnan kanssa kukaties eläneet kolmekymmentä vuotta ja jättäneet jälkeläisillenne rikkaan viinitarhan ja tuhannen desjatiinaa maissipeltoa, jota vastoin minä tunsin itseni hävinneeksi mieheksi jo ensi päivästä lähtien. Kaupungissa on sietämätön helle, ikävää ja ihmistyhjää; jos taas lähtee kaupungin ulkopuolelle, on siellä joka pensaan ja kiven juurella uhkaamassa skorpionit ja käärmeet, ja taas edempänä ovat vuoret ja erämaa. Outoja ovat ihmiset, outoa on luonto, viljelys kurjaa — kaikki se on jotakin toista kuin turkki päällä kävellä Nevskillä käsi kädessä Nadeshda Feodorovnan kanssa ja haaveksia lämpimistä maista. Se kysyy taistelua elämästä ja kuolemasta, mutta mikä taistelija minä olen? Kurja, heikkohermoinen ja työhön tottumaton… Ensi päivästä asti oivalsin, että raatajan elämää ja viinitarhaa koskevat ajatukseni olivat pötyä. Mitä taas tulee rakkauteen, täytyy minun sanoa, että elämä naisen kanssa, joka on lukenut Spenceriä ja sinun tähtesi matkannut maailman ääriin, on yhtä vähän mielenkiintoista kuin
- 67. elämä minkä Anisan tai Akulinan kanssa hyvänsä. Sama silitysraudan, ihojauheen ja lääkkeiden haju, samat paperikähertimet joka ikinen aamu ja sama itsepetos… — Silitysraudatta ei taloudessa tulla toimeen, — huomautti Samoilenko, punastuen siitä, että Lajevski niin avomielisesti puheli hänelle tutusta naisesta. — Sinä, Vanja, olet tänään pahalla tuulella, huomaan… Nadeshda Feodorovna on erinomainen, sivistynyt nainen, sinä mainion järkevä mies. Tosin ette ole vihityt, — sanoi Samoilenko, katsahtaen hämillään pöydälle päin, — mutta sehän ei ole teidän vikanne, ja sitäpaitsi … pitää osata vapautua ennakkoluuloista ja asettua nykyajan kannalle… Minä itse puolustan siviiliavioliittoa, ja … mutta olen sitä mieltä, että jos kerran olette yhteen käyneet, on teidän yhdessä elettävä kuolemaanne asti. — Ilman rakkauttako? — Selitän sinulle heti, — sanoi Samoilenko. — Noin kahdeksan vuotta takaperin meillä oli täällä asioitsijana vanha ukko, ylen viisas mies. Hänellä oli tapana sanoa: perhe-elämässä on tärkeintä kärsivällisyys. Kuuletko, Vanja? Ei rakkaus, vaan kärsivällisyys… Elettyäsi parisen vuotta rakkaudessa on perhe-elämäsi silminnähtävästi joutunut sellaiseen vaiheeseen, jolloin sinun, voidaksesi ylläpitää tasapainoa, on pantava liikkeelle kaikki kärsivällisyytesi… — Sinä uskot asioitsijasi, tuon rahjuksen lausuntoon. Minusta hänen neuvonsa sensijaan on mielettömyyttä. Sinun äijäsi saattoi teeskennellä, saattoi harjoitella kärsivällisyyttä, jolloin se henkilö, johon hän ei tuntenut rakkautta, esiintyi hänelle kärsivällisyyden harjoituksissa välttämättömänä esineenä, mutta minä en ole vielä niin syvälle vajonnut; jos minua miellyttää kärsivällisyyttä harjoitella,
- 68. ostan itselleni voimistelupainot tahi vikurin hevosen, mutta ihmisen jätän rauhaan. Samoilenko tilasi valkoista jäänsekaista viiniä. Kun oli juotu lasi mieheen, kysäisi Lajevski äkkiä: — Sanoppas, ole niin hyvä, mitä merkitsee aivojen pehmeneminen? — Se on, kuinka sinulle selittäisin … taudillinen tila, jossa aivoaine pehmenee … aivan kuin vetelöityy. — Voiko sitä parantaa? — Voi kyllä, ellei tauti ole saanut kehittyä varsin pitkälle… Kylmiä suihkukylpyjä, espanjankärpäslaastaria… Jotakin sisällisesti nautittavaksi. — Vai niin… Näetkös siis, mimmoinen on tilani. Hänen kanssaan en voi elää: se käy yli voimieni. Niin kauan kuin olen sinun seurassasi, minä vielä sekä filosofoin että naureskelen, mutta kotona mieleni kokonaan lamaantuu. On niin äärettömän tukala ollakseni, että jos minulle sanottaisiin esimerkiksi, että minun on vielä elettävä hänen kanssaan yksi kuukausi, niin on luultavaa, että laskisin luodin otsaani. Siitä huolimatta minun on mahdoton hänestä erota… Hän on yksinäinen, ei osaa tehdä työtä, rahoja ei ole minulla eikä hänellä… Minne hän joutuu? Kenenkä luo menisi? En voi keksiä mitään … Sano nyt: mitä minun on tekeminen? — Tjaa … murahti Samoilenko, tietämättä mitä vastata. — Rakastaako hän sinua?
- 69. — Rakastaa, sen verran kuin mies hänen ikäiselleen ja luonteiselleen naiselle on tarpeellinen. Minusta hänen olisi yhtä vaikea erota kuin ihojauheestaan ja paperikähertimistään. Minä olen hänen naiskammionsa välttämätön, olennainen osa. Samoilenko joutui hämille. — Sinä olet, Vanja, tänään pahalla päällä, — sanoi hän. — Et tainnut nukkua. — Sepä se, nukuin huonosti… Tunnen yleensä voivani pahoin. Päässä tyhjyys, sydäntä ahdistaa, jokin yleinen heikkous… Ihan täytyy lähteä karkuun! — Minne? — Sinne, pohjolaan. Mäntyjen, sienien, ihmisten, aatteiden pariin… Antaisin puolet elämästäni, jos tällä hetkellä saisin jossakin Moskovan tai Tulan kuvernementissa uida joessa, palella, ymmärräthän, kuljeskella sitten kolmisen tuntia vaikka kaikkein mitättömimmän ylioppilaan kanssa ja lörpötellä, lörpötellä… Entä heinä, — kuinka se tuoksuu! Muistatko? Iltaisin kun käyskentelee puistossa, sisältä kuuluu pianonsoittoa, kuulee junan tulevan… Mielihyvä saattoi Lajevskin naurahtamaan, hänen silmistään herahtivat kyynelet, ja salatakseen ne hän paikaltaan nousematta kurottautui ottamaan tulitikkuja viereiseltä pöydältä. — Minäpä en ole kahdeksaantoista vuoteen käynyt Venäjällä, — sanoi Samoilenko. — Olen unohtanut, millaista siellä onkaan. Mielestäni ei ole ihanampaa maata kuin Kaukaasia.
- 70. — On eräs Vereshtshaginin taulu: mahdottoman syvän kaivon pohjalla nääntyvät muutamat hengiltä tuomitut vaivoihinsa. Juuri semmoiselta syvältä kaivolta minusta näyttää ihana Kaukaasiasi. Jos minun olisi valittava kahdesta: olla nokikolarina Pietarissa tahi täkäläisenä ruhtinaana, niin valitsisin edellisen toimen. Lajevski vaipui mietteisiin. Silmäillessään hänen kumaraista vartaloaan, yhteen kohti tähystäviä silmiään, kalpeita, hikisiä kasvojaan ja kuopalle painuneita ohimottaan, pureskeltuja kynsiä ja tohvelia, joka riippuen kantapäästä lonkollaan jätti huonosti parsitun sukan näkyviin, Samoilenkon tuli häntä sääli, ja kaiketikin sentähden, että Lajevski muistutti hänen mielestään avutonta lasta, hän kysyi: — Elääkö äitisi? — Elää, mutta me olemme huonoissa väleissä. Hän ei voinut antaa minulle anteeksi tätä rakkausjuttua. Samoilenko oli hartaasti mieltynyt ystäväänsä. Hän näki Lajevskissa hulivilin, ylioppilaan, hyvälaatuisen nuoren miehen, jonka kanssa sopi sekä ryypätä että nauraa makeasti ja haastaa suoraan sydämestä. Se, mitä hän käsitti Lajevskista, ei häntä ollenkaan miellyttänyt. Lajevski naukkaili aika lailla, silloinkin kun ei olisi sopinut, pelasi korttia, halveksi virkaansa, eli yli varojensa, käytti puheessaan usein säädyttömiä lausetapoja, kulki kadulla tohvelit jalassa ja riiteli syrjäisten kuullen Nadeshda Feodorovnan kanssa — ja se oli Samoilenkolle vastenmielistä. Mutta se, että Lajevski oli opiskellut filologisessa tiedekunnassa, tilasi kahta paksua aikakauskirjaa, puhui usein niin oppineesti, että vain harvat häntä ymmärsivät, vietti yhdyselämää sivistyneen naisen kanssa — tätä kaikkea Samoilenko ei ymmärtänyt, mutta se miellytti häntä, ja hän piti Lajevskia suuressa arvossa, itseänsä ylempänä.
- 71. — Vielä yksi seikka, — sanoi Lajevski pudistaen päätään. — Mutta se olkoon vain meidän kesken… En ole ilmaissut sitä Nadeshda Feodorovnallekaan, joten muista pitää hampaittesi takana… Sain toissa päivänä kirjeen, jossa ilmoitettiin hänen miehensä kuolleen aivojen pehmenemiseen. — Hyväinen aika … huokasi Samoilenko. — Miksi sitä häneltä salaat? — Kirjeen näyttäminen hänelle merkitsisi samaa kuin lähde pois kirkkoon vihille. Mutta ensin ovat välimme suoritettavat. Sittenkun hän on tullut huomaamaan, että yhdyselämämme jatkuminen on mahdotonta, näytän hänelle kirjeen. Silloin se ei enää ole vaarallista. — Tiedätkö mitä, Vanja? — sanoi Samoilenko, ja hänen kasvoilleen tuli äkkiä surullinen, rukoileva ilme, kuin hän olisi ollut aikeissa pyytää jotakin hyvin suloista ja pelännyt saavansa kieltävän vastauksen. — Nai sinä hänet. — Minkä tähden? — Täytä velvollisuutesi tätä viehättävää naista kohtaan! Hänen miehensä kuoli pois, joten siis itse kaitselmus ikäänkuin viittaa sinulle, mitä on tehtävä! — Kummallinen ihminen, pitäisihän sinun käsittää, että se on mahdotonta. Yhtä halpamaista ja ihmisarvoa alentavaa on naida ilman rakkautta kuin toimittaa jumalanpalvelusta, jos puuttuu uskoa. — Mutta se on velvollisuutesi! — Miksi minä olisin velvollinen? — kysyi Lajevski kiihtyneenä.
- 72. — Siksi, että sinä houkuttelit hänet miehensä luota ja otit hänet vastuullesi. — Olenhan sanonut jo ihan selvästi: minä en rakasta! — Kun et rakastane, niin pidä kunniassa, palvo… — Pidä kunniassa, palvo … matki Lajevski. — Kuin mitäkin abbedissaa… Sinä olet huono psykologi ja fysiologi, jos luulet, että naisen kanssa elääksesi selviydyt yksistään kunnioittamisella ja palvomisella. Naiselle on ennen kaikkea tarpeen makuuhuone. — Vanja, Vanja … nuhteli Samoilenko hämillään. — Sinä olet vanha lapsi, teoreetikko, minä sen sijaan nuori ukko ja käytännön mies, me emme milloinkaan opi ymmärtämään toisiamme. Parasta kun lopetamme koko keskustelun… Mustaa! — huusi Lajevski viinurille, — paljonko olemme velkaa? — Ei, ei, — virkkoi tohtori hätäisesti ja tarttui Lajevskin käsivarteen. — Minä maksan tämän. Minähän tilasinkin. Merkitse minun laskuuni! — huusi hän Mustaalle. Ystävykset nousivat ja lähtivät äänettöminä astumaan rantakatua pitkin. Puistikkokadulle kääntyessään he pysähtyivät ja puristivat jäähyväisiksi toistensa kättä. — Te, herrat, olette liiaksi hemmoiteltuja! — huokasi Samoilenko. — Sinulle on kohtalo lähettänyt nuoren, kauniin, sivistyneen naisen, ja sinä pyrit hänestä eroon; jos minulle Jumala sen sijaan antaisi vaikka kierokylkisen mummon, kunhan olisi lempeä ja hyväluontoinen, niin olisinpa onnellinen! Eläisin hänen kanssaan viinitarhassani ja…
- 73. Samoilenko huomasi viimeisen lausumansa sopimattomaksi ja lisäsi: — Hommailisihan se velho siellä edes teekeittiötä kiehumaan! Heitettyään Lajevskille hyvästit hän lähti astumaan puistikkokatua pitkin. Kun hän jykevänä, kasvoilla ankara sävy, yllään lumivalkoinen palttinamekko, jalassa huolellisesti kiilloitetut saappaat, röyhistäen rintaansa, jolla koreili Vladimirin tähti nauhoineen, asteli pitkin katua, oli hän hyvin mieltynyt omaan itseensä ja hänestä näytti, että koko maailma mielihyvin katseli häntä. Päätään kääntämättä hän katsoi sivuilleen ja huomasi, että puistikkokatu oli hyvässä kunnossa, että nuoret kypressit, kumipuut ja rumat, kuivankituliaat palmut olivat sangen kauniita ja lupasivat vastaisuudessa levittää taajaa varjoa, että tserkessit ovat rehellistä, vieraanvaraista kansaa. Kumma, ettei Lajevskia miellytä Kaukaasia, ajatteli hän, varsin kummallista. Samassa tuli vastaan viisi sotamiestä pyssyt olalla, tehden hänelle kunniaa. Oikealla puolen puistikkokatua kulki jalkakäytävää myöten erään virkamiehen rouva poikansa, lukiolaisen, seurassa. — Maria Konstantinovna, hyvää huomenta! — huusi hänelle Samoilenko, herttaisesti hymyillen. — Kävittekö uimassa? Ha-ha- haa… Terveiseni Nikodim Aleksandritshille! Ja hän asteli edelleen, hymyillen yhä herttaisesti; mutta nähdessään sotilasvälskärin tulevan vastaan hän äkkiä rypisti kulmansa, pysäytti välskärin ja kysyi: — Onko sairaalassa ketään? — Ei ketään, teidän ylhäisyytenne.
- 74. — Kuinka? — Ei ketään, teidän ylhäisyytenne. — Hyvä, saat mennä… Huojutellen mahtavasti ruumistaan hän suuntasi askelensa vesimyymälälle päin, jossa pöydän takana istui grusialaiseksi itseään mainitseva vanhahko, lihava juutalaisakka, ja lausui tälle niin kovaa, kuin olisi komentanut kokonaista rykmenttiä: — Olkaa hyvä, antakaa minulle soodavettä!
- 75. II Lajevskin lemmetön suhde Nadeshda Feodorovnaan ilmeni etupäässä siinä, että kaikki, mitä tämä sanoi tai teki, tuntui hänestä valheelta tai valheeseen vivahtavalta ja että kaikki naisia ja rakkautta vastaan tähdätyt kirjoitukset, jotka hän oli lukenut, näyttivät hänestä mahdollisimman parhaiten soveltuvan häneen, Nadeshda Feodorovnaan, ja tämän mieheen. Lajevskin palatessa kotiin hän istui jo täysin puettuna ja kammattuna ikkunan ääressä, joi huolestuneen näköisenä kahvia ja selaili paksua aikakauskirjaa; ja Lajevski ajatteli, ettei kahvinjuonti ole mikään niin merkillinen tapahtuma, että sen vuoksi kannattaisi tekeytyä huolestuneen näköiseksi, ja että turhaan hän oli kuluttanut aikaa taiteellisen hiuslaitteen aikaansaamiseksi, koska täällä ei ollut ketään, jota tulisi koettaa miellyttää. Aikakauskirjassakin hän havaitsi valhetta. Hän ajatteli, että Nadeshda Feodorovna pukeutui ja piti huolta hiuslaitteestaan näyttääkseen kauniilta ja luki näyttääkseen viisaalta. — Mitä arvelet, sopisiko minun tänään mennä uimaan? — kysyi Nadeshda Feodorovna.
- 76. — Mitäs siitä? Menet tahi olet menemättä, ei tuosta luulisi maanjäristystä seuraavan… — Ei, minä kysyn vain sentähden, että kunhan ei tohtori suuttuisi. — Kysy sitten tohtorilta. Minä en ole tohtori. Tällä kertaa tuntuivat Lajevskista kaikkein vastenmielisimmiltä Nadeshda Feodorovnassa hänen valkoinen avoin kaulansa ja hiuskiharat niskassa, ja hän muisti, että Anna Kareninaa, kun hänen rakkautensa miestänsä kohtaan kylmeni, eivät miellyttäneet hänen korvansa, ja ajatteli: Kuinka oikein se on! Kuinka oikein! Tuntien heikkoutta sekä päässään tyhjyyttä hän meni työhuoneeseensa, kävi sohvalle pitkäkseen ja peitti kasvonsa nenäliinalla, etteivät kärpäset pääsisi tekemään kiusaa. Velton unteloja ajatuksia aina yhdestä ja samasta asiasta liikkui hänen aivoissaan kuin pitkä kuormajono sateisena syysiltana, ja hän vaipui alakuloiseen unenhouraukseen. Hän tunsi olevansa syyllinen Nadeshda Feodorovnan ja tämän miehen edessä, jopa siihenkin, että viimemainittu oli kuollut. Hän tunsi olevansa syyllinen oman elämänsä edessä, jonka oli turmellut, korkeiden aatteiden, tietojen ja työn maailman edessä, ja tämä ihmeellinen maailma tuntui hänestä mahdolliselta ja olemassa olevalta, ei täällä rannikolla, missä nälkäiset turkkilaiset ja laiskat abhasialaiset kuljeskelevat, vaan siellä, pohjolassa, missä ovat ooppera, teatterit, sanomalehdet ja kaikki henkisen työn muodot. Rehellinen, viisas, ylevä ja puhdas saattaa olla ainoasti siellä, ei täällä. Hän syytti itseään siitä, ettei hänellä ollut ihanteita eikä elämää johtavaa aatetta, vaikka nyt hämärästi tajusi, mitä se merkitsi. Kaksi vuotta takaperin, kun hän oli rakastunut Nadeshda Feodorovnaan, hänestä tuntui, että heti kun hän on päässyt yhdyselämään Nadeshda Feodorovnan kanssa ja he ovat
- 77. matkustaneet Kaukaasiaan, niin hän on pelastettu alennuksesta ja elämän tyhjyydestä; samaten hän nytkin oli varma, ettei muuta tarvita kuin että hän hylkää Nadeshda Feodorovnan ja lähtee Pietariin, niin hän saa kaikki, mitä kaipaa. — Pakoon! — murahti hän, nousten istualleen ja pureksien kynsiään. — Pakoon! Hänen mielikuvituksensa hahmotteli, kuinka hän astuu laivaan ja sitten syö aamiaista, juo kylmää olutta, puhelee laivan kannella naisten kanssa, sitten Sevastopolissa nousee junaan ja antaa huhkia. Terve, vapaus! Asemat vilahtavat ohi toinen toisensa jälkeen, ilma käy yhä kylmemmäksi ja kalseammaksi, jo näkyy koivuja ja kuusia, tuossa jo Kursk ja Moskova… Ravintoloissa kaalikeittoa, lampaanlihaa ja puuroa, samminlihaa, olutta, sanalla sanoen — ei enää aasialaista raakalaisuutta, vaan Venäjä, todellinen Venäjä. Matkustajat junassa puhuvat kaupasta, uusista laulajista, ranskalaisvenäläisistä ystävyyssuhteista; kaikkialla on jo tunnettavissa virkeä, älykäs, reipas sivistyselämä… Pikemmin, pikemmin! Jopa ollaan vihdoin Nevskillä, Isolla Morskajalla, tuossa on Kovnon solakatukin, missä hän muinoin asui ylioppilaitten kanssa yhdessä, tuossa tuttu harmaja taivas, tihkusade, märät ajurit… — Ivan Andreitsh! — kutsui joku viereisestä huoneesta. — Oletteko kotona? — Täällä olen, — vastasi Lajevski. — Mitä tahdotte? — Täällä olisi asiapapereita!
- 78. Lajevski nousi vastahakoisesti, tunsi päätänsä huimaavan ja lähti haukottaen ja tohveleillaan laahustaen viereiseen huoneeseen. Siellä seisoi kadulla avonaisen ikkunan edessä eräs hänen nuoria virkatovereitaan ja levitteli ikkunalaudalle virkakirjeitä. — Kohtsillään, ystäväiseni, — sanoi Lajevski suopeasti ja lähti etsimään musteastiaa; lähestyttyään jälleen ikkunaa hän lukematta allekirjoitti paperit ja virkkoi: — Kyllä on kuuma! — Kuuma on. Tuletteko tänään? — Tuskinpa… En ole oikein terve. Sanokaa, ystävä hyvä, Sheshkovskille, että puolisen jäljestä pistäydyn hänen luonaan… Virkamies lähti pois. Lajevski kävi jälleen sohvalle pitkäkseen ja alkoi miettiä. Siis on tarkoin punnittava kaikki asianhaarat ja harkittava… Ennenkuin lähden täältä, on minun suoritettava velkani. Olen velkaa noin kaksituhatta ruplaa. Rahaa minulla ei ole… Se tietysti ei ole tärkeää: osan suoritan jo nyt miten kuten, osan lähetän sitten Pietarista… Tärkeintä on Nadeshda Feodorovna… Ihan ensiksi täytyy saada meidän välimme selviksi… Niin. Tovin kuluttua hän aprikoi: eikö olisi parasta lähteä Samoilenkon kanssa neuvottelemaan? Voisi tuota lähteä, ajatteli hän, mutta mitä siitä on hyötyä? Rupean taas syyttä suotta puhumaan hänelle naiskammiosta, naisista ja siitä, mikä on kunniallista, mikä ei. Onko, lempo soikoon, paikallaan puhua kunniallisuudesta tai epäkunniallisuudesta, jos on
- 79. kysymyksessä saada mahdollisimman pian pelastetuksi oma elämäni, jos menehdyn tässä hemmetin orjuudessa ja teen lopun itsestäni?… Täytyyhän vihdoinkin ymmärtää, että tämmöisen elämän jatkaminen kuin minun on halpamaista ja julmaa, jonka rinnalla kaikki muu on vähäpätöistä ja arvotonta. Pakoon! mutisi hän, nousten istualleen. Pakoon! Autio merenranta, hellittämätön kuumuus, yksitoikkoiset, punasinervät, autereiset vuoret, aina yhtäläiset ja synkät, ikävystyttivät häntä ja tuntuivat uuvuttavan hänet uneen ja unholaan. Saattaa olla, että hän on hyvin järkevä, lahjakas ja erinomaisen rehellinen; saattaa olla, että jolleivät häntä joka puolelta olisi ympäröineet meri ja vuoret, hänestä olisi voinut sukeutua joku erinomainen toimihenkilö kunnalliselämän alalla, valtiomies, puhuja, sanomakirjailija, aatteiden esitaistelija. Kenpä tietää! Näin ollen, eikö ole järjetöntä kiistellä siitä, onko kunniallista vai epäkunniallista, jos lahjakas ja hyödyllinen ihminen, esimerkiksi soittotaituri tai taidemaalari, karatakseen vankeudesta, rikkoo seinän ja pettää vartijansa? Semmoisen ihmisen asemassa kaikki on kunniallista. Kello kaksi Lajevski ja Nadeshda Feodorovna istuutuivat syömään päivällistä. Palvelijan tuotua pöytään riisikeittoa tomaattien kanssa, sanoi Lajevski: — Joka päivä aina samaa. Miksi ei keitetä kaalisoppaa? — Ei ole kaalia. — Kummallista. Samoilenkolla keittävät kaalisoppaa, ja Maria Konstantinovnalla keitetään kaalia. En tiedä, minkä tähden minun yksistään on aina syötävä tätä imelää sotkua. Eihän käy laatuun sillä tavalla, kyyhkyseni.
- 80. Kuten aviopuolisojen suuren suurella enemmistöllä ei Lajevskilla ja Nadeshda Feodorovnalla aikaisemmin saatu yhtäkään päivällistä syödyksi ilman oikkuja ja kinastelua, muita siitä pitäen, kun Lajevski oli päättänyt, ettei hän enää rakasta, koetti hän kaikessa antaa Nadeshda Feodorovnalle myöten, puhutteli häntä suopeasti ja kohteliaasti, hymyili, nimitti kyyhkysekseen. — Tällä liemellä on lakritsan maku, — sanoi hän naurahtaen, hän koetti väkisinkin näyttää ystävälliseltä, mutta ei voinut, ja lausui: — Meillä ei kukaan pidä taloudesta huolta… Jos sinä tosiaankin olet niin sairas tai lukemisesi sinua estää, niin otan minä hoitaakseni keittiöpuolta. Ennen Nadeshda Feodorovna olisi hänelle vastannut: ota vaan, tai: tahdot näköjään tehdä minusta piian, mutta nyt hän vain arasti katsahti Lajevskiin ja punastui. — No, kuinka jaksat tänään? — kysyi Lajevski ystävällisesti. — Eipä valittamista. Tunnen vain itseäni raukaisevan. — Täytyy pitää vaaria terveydestään, kyyhkyseni. Minä suorastaan pelkään puolestasi. Nadeshda Feodorovnalla oli jokin tauti. Samoilenko sanoi hänellä olevan vuorottelevan kuumetaudin ja määräsi hänelle kiniiniä; sen sijaan toinen lääkäri, Ustimovitsh, pitkäkasvuinen, laihahko, ihmisiä karttava mies, joka päivällä istui kotonaan, mutta iltaisin käveli hiljakseen yskien pitkin rantakatua, kädet selän takana ja kävelykeppi sojottaen pitkin selkää, väitti hänellä olevan naistentaudin ja määräsi kylmiä kääreitä. Siihen aikaan, jolloin Lajevski vielä rakasti, herätti Nadeshda Feodorovnan tauti hänessä
- 81. sääliä ja pelkoa, mutta nyt hän tuossa taudissakin huomasi valhetta. Kellervät, uneliaat kasvot, veltto katse ja haukottelut, jotka seurasivat Nadeshda Feodorovnan vilutaudinkohtauksia, ynnä se, että hän taudinpuuskan aikana maatessaan piti yllään matkapeitettä ja oli enemmän pojan kuin naisen näköinen, ja että hänen huoneessaan oli tukahuttava ilma ja omituisen vastenmielinen haju — kaikki tämä oli Lajevskin mielestä omiaan hälventämään onnenkuvitelmaa, samalla kun se oli voimakas vastalause rakkautta ja avioliittoa vastaan. Toisena ruokalajina hänelle tarjottiin pinaattia ja koviksi keitettyjä munia, mutta Nadeshda Feodorovnalle, kuten ainakin sairaalle, marjahyytelöä ja maitoa. Kun Nadeshda Feodorovna huolestuneen näköisenä ensin kosketti lusikalla hyytelöä ja sitten ryhtyi sitä laiskasti syömään, ryypäten maitoa päälle, joutui Lajevski kuullessaan kulahduksia hänen kurkussaan semmoisen vihan valtaan, että hänellä ihan alkoi pää syhyä. Hän myönsi kyllä, että tuommoinen tunne olisi koiraakin vastaan loukkaava, mutta ei kiukustunut itseensä, vaan Nadeshda Feodorovnaan siitä, että tämä herätti hänessä sen tunteen, ja ymmärsi nyt, miksi rakastajat toisinaan surmaavat rakastettunsa. Itse hän ei aikonut murhata, mutta valamiesoikeuden jäsenenä hän olisi julistanut murhamiehen syytteestä vapaaksi. — Kiitos, kyyhkyseni, — sanoi hän päivälliseltä noustua ja suuteli Nadeshda Feodorovnaa otsaan. Tultuaan työhuoneeseen hän asteli noin viisi minuuttia nurkasta nurkkaan, katseli syrjäkulmin saappaitaan, istui sitten sohvaan ja jupisi: — Pakoon, pakoon! Välit selviksi ja pakoon!
- 82. Hän kävi pitkäkseen sohvalle, ja hänelle muistui taas, että hän kenties oli aiheuttanut Nadeshda Feodorovnan miehen kuoleman. — Syyttää ihmistä siitä, että hän on rakastunut tahi lakannut rakastamasta, on typerää, — uskotteli hän itselleen ja koukisti pitkällään ollen säärensä saadakseen saappaat vedetyiksi jalkaan. — Rakkaus ja viha eivät ole meidän käskettävissämme. Mitä taas tulee aviomieheen, niin kenties minä välillisesti olenkin ollut yhtenä syynä hänen kuolemaansa, mutta toiseksi, olenko minä sittenkään syypää siihen, että rakastuin hänen vaimoonsa ja vaimo minuun? Sitten hän nousi, otti lakkinsa ja läksi virkatoverinsa Sheshkovskin luo. Sinne kokoontui joka päivä virkamiehiä pelaamaan vistiä ja juomaan kylmää olutta. Epäröimiselläni minä muistutan Hamletia, ajatteli Lajevski matkalla. Kuinka kohdalleen Shakespeare on osannut!
- 83. III. Haihduttaakseen ikävää ja suopeasti suhtautuen vasta saapuneiden sekä perheettömien tulokasten tukalaan tilaan, näillä kun ei ollut missä syödä päivällistä — kaupungissa ei näet ollut ravintolaa — tohtori Samoilenko ylläpiti kotonaan jonkinlaista yhteistä päivällispöytää. Nyt puheenaolevaan aikaan hänellä oli vain kaksi ruokavierasta: nuori eläintieteilijä von Coren, joka oli kesällä saapunut Mustalle merelle tutkimaan meduusojen sikiökehitystä, ja diakoni Pobedov, joka oli äskettäin päässyt pappisseminaarista ja määrätty tähän pikkukaupunkiin toimittamaan terveyttänsä hoitamaan matkustaneen vanhan apupapin virkaa. He maksoivat kumpikin kaksitoista ruplaa kuussa päivällisestä ja illallisesta, ja Samoilenko oli vaatinut heitä kunniasanallaan lupaamaan, että he täsmälleen kello kaksi saapuisivat päivälliselle. Ensimmäisenä saapui tavallisesti von Coren. Hän istuutui äänetönnä vierashuoneeseen ja ottaen pöydältä albumin ryhtyi tarkkaavaisesti katselemaan himmentyneitä valokuvia, jotka esittivät joitakin tuntemattomia mieshenkilöitä, yllään leveät housut ja päässä silkkihatut, sekä myssypäisiä naisia, joilla oli pönkkähameet; vain muutamat näistä Samoilenko muisti nimeltä, mutta niistä, jotka hän oli unohtanut, hän virkkoi huoahtaen: Erinomainen, tuiki viisas
- 84. mies! Lopetettuaan albumin tarkastelun von Coren otti hyllyltä pistoolin ja puristaen kiinni vasemman silmänsä tähtäsi hyvän aikaa ruhtinas Vorontsovin muotokuvaan tahi katseli kuvastimesta tummanvereviä kasvojaan, leveää otsaansa ja mustaa, kähärää neekeritukkaansa ja himmeästä, suurikukkaisesta, persialaista mattoa muistuttavasta pumpulikankaasta tehtyä paitaansa ja leveää nahkavyötään, joka hänellä oli liivien asemesta. Itsensä tarkasteleminen tuotti hänelle miltei suurempaa mielihyvää kuin valokuvien tai kallishelaisen pistoolin katseleminen. Hän oli hyvin tyytyväinen sekä kasvoihinsa että huolellisesti kerittyyn partaansa ja leveihin hartioihinsa, jotka olivat ilmeisenä todistuksena hyvästä terveydestä ja vankasta ruumiinrakenteesta. Samoin hän oli tyytyväinen keikarimaiseen pukuunsa, alkaen kaulaliinasta, joka oli paidan väriin mukautuvasti valittu, keltaisiin kenkiinsä asti. Sillä välin, kun hän katseli albumia ja seisoi kuvastimen edessä, pidettiin keittiössä ja sen eteen aukeavassa porstuassa kovaa mekastusta ja kiirettä; siellä näet itse Samoilenko, takitta ja liiveittä, rinta avoinna, ärtyisenä ja hikeä valuen touhusi pöytien ääressä valmistaen salaattia tahi jotakin kastiketta, tahi lihaa, kurkkuja ja sipulia erästä suosittua okroshka-nimistä ruokalajia varten, ja sitä tehdessään hän sinkautteli mulkoilevia katseita avustavaan sotamiespalvelijaan, väliin taas huitaisi tätä kohden milloin veitsellä, milloin kauhalla. — Etikka tänne! — komensi hän, — tahi ei etikkaa, vaan ruokaöljyä! — tiuskasi hän, polkien jalkaa. — No, minne sinä katosit, mölhö? — Lähdin noutamaan voita, teidän ylhäisyytenne, — vastasi hätääntyneesti palvelija särkyneellä tenoriäänellä.
- 85. — Pian nyt! voi on kaapissa! Ja sano Darjalle, että hän lisää kurkkupönttöön tilliä! Tilliä! Peitä hapankerma-astia, etteivät kärpäset mene sinne, tolvana! Ja tuntui kuin koko talo olisi ollut täynnä hänen huutonsa pauhua. Noin kymmentä tai viittätoista minuuttia ennen kello kahta saapui diakoni, nuori kahdenkolmatta ikäinen mies, laihahko, parraton, viikset vasta hieman oraalla. Astuessaan vierashuoneeseen hän teki pyhimyskuvaa kohden kääntyen ristinmerkin ja ojensi hymyillen von Corenille kättä. — Päivää, — virkkoi eläintieteilijä kylmästi. — Missä olette ollut? — Rantalaiturilla merihärkiä onkimassa. — Tietysti, missäs… Teidän, diakoni, näyttää olevan mahdoton koskaan ryhtyä vakavampaan työhön. — Kuinka niin? Eihän työ ole karhu, jotta metsään pakenisi, — vastasi diakoni hymyillen ja työnsi kätensä syvälle valkoisen pappisviittansa taskuihin. — Selkään sietäisitte saada! — huokasi eläintieteilijä. Kului vielä noin neljännestunti, mutta syömään ei kutsuttu, kuului vain, kuinka sotamies juosta paukutteli porstuasta keittiöön ja taas takaisin, koluten saappaillaan, ja Samoilenko huusi: — Älä sinne tuppaa! Pane pöydälle! Pese ensin! Nälissään diakoni ja von Coren alkoivat kengänkoroillaan jyskyttää lattiaa, siten ilmaisten kärsimättömyyttään, kuten toisen rivin yleisö teatterissa. Viimein avattiin ovi, ja sotamiespalvelija ilmoitti peräti väsyneenä: Saisi tulla ruualle. Ruokasalissa he kohtasivat Samoilenkon, keittiön
- 86. kuumuudesta naama punaisena; hän katsoi heihin tuimasti ja nostaen kauhun ilme kasvoillaan kannen liemimaljasta pani heille kummallekin lautasellisen keittoa, ja vasta sitten, kun hän oli varmistunut siitä, että he söivät hyvällä halulla ja että ruoka heille maistui, päästi hän helpotuksen huokauksen ja istui syvään nojatuoliinsa. Hänen ilmeensä lieveni ja hän kaatoi itselleen hitaasti ryypyn viinaa ja lausui: — Nuoren polven terveydeksi! Lajevskin kanssa tapahtuneen keskustelun jälkeen Samoilenko oli koko ajan aamusta puolipäivään asti, vaikka olikin erittäin hyvällä tuulella, tuntenut sielunsa syvyydessä jonkinlaista painostusta. Hänen oli sääli Lajevskia ja halutti auttaa häntä. Otettuaan liemiruuan alle ryypyn hän huokasi ja sanoi: — Tapasin tänään Vanja Lajevskin. Raskasta on miehen elämä. Aineellinen puoli ei ole kehuttava, mutta pahempaa on, että hän henkisesti on aivan masentunut. Sääli sitä poikaa! — Minun vain ei ole häntä sääli! — sanoi von Coren. — Jos tuo miellyttävä mies olisi hukkumaisillaan, niin minä vielä kepillä sysäisin: huku, veikkonen, huku… — Ei ole totta. Sitä et sinä tekisi. — Mistä sen päätät? — kysyi eläintieteilijä olkapäitään kohauttaen. — Kykenen minä tekemään hyväntyön siinä missä toinenkin. — Ihmisen hukuttaminen, sekö olisi hyvätyö? — virkkoi diakoni naurahtaen.
- 87. — Ainakin silloin, kun on puhe Lajevskista. — Liemestä puuttuu jotakin, — lausui Samoilenko, kääntääkseen puheen toiselle tolalle. — Lajevski on ehdottomasti vahingollinen ja yhteiskunnalle yhtä vaarallinen kuin koleeramikroobi, — jatkoi von Coren. — Kelpo työn tekee, ken hänet hukuttaa. — Ei ole sinulle kunniaksi, että puhut lähimmäisestäsi tuolla tavalla. Sanoppa, miksi sinä häntä vihaat? — Älä puhu, tohtori, joutavia. Tietysti olisi järjetöntä vihata ja halveksia mikroobia, mutta pitää kaiken uhalla lähimmäisenään ketä hyvänsä ilman erotusta on samaa kuin olla kokonaan harkitsematta, kieltäytyä oikeudenmukaisesti suhtautumasta ihmisiin, sanalla sanoen — pestä kätensä. Minä pidän Lajevskia heittiönä, sitä en salaa, ja kohtelen häntä kuten heittiötä, täysin tunnollisesti: sinä pidät häntä lähimmäisenäsi, no — suutele siis häntä; pidät lähimmäisenä, mikä tietää, että kohtelet häntä samalla tavalla kuin minua ja diakonia, se on — et niin etkä näin. Sinä olet yhtä piittaamaton kaikkia kohtaan. — Että voikin sanoa ihmistä heittiöksi! — murahti Samoilenko ja rypisti inhoa ilmaisten kulmiaan. — Se on siinä määrin epähienoa, etten osaa sanoakaan! — Ihmisiä arvostellaan heidän tekojensa mukaan, — pitkitti von Coren. — Päättäkää te, diakoni… Puhun, diakoni, nyt teille. Herra Lajevskin toiminta on levitetty julkisesti eteenne kuin pitkä kiinalainen asiakirja, ja te voitte lukea sen alusta loppuun. Mitä hän
- 88. on tehnyt niiden kahden vuoden aikana, jotka hän on oleskellut täällä? Ensiksi hän on opettanut kaupungin asukkaat pelaamaan vistiä. Kaksi vuotta takaperin se peli oli täällä tuntematon; nyt sitä pelaavat aamusta myöhään yöhön kaikki, yksin naiset ja keskenkasvuiset nuorukaisetkin. Toiseksi hän on opettanut seudun asukkaat juomaan olutta, mikä myöskin oli tuntematonta täällä; häntä saavat asukkaat kiittää siitä, että heillä on tietoja eri viinalajeista, ja nyt he osaavat silmät ummessa erottaa Koshelevin viinan Smirnovin n:o 21:stä. Kolmanneksi, ennen täällä pidettiin yhteyttä toisten miesten vaimojen kanssa salaa, samoista syistä kuin varkaatkin harjoittavat ammattiaan salassa eikä julkisesti; huorintekoa pidettiin jonakin semmoisena, mitä hävettiin tuoda julkisesti näytteille; mutta Lajevski esiintyi tässäkin suhteessa uranuurtajana: hän viettää yhdyselämää toisen miehen vaimon kanssa julkisesti. Neljänneksi… Von Coren söi sukkelasti loppuun keiton ja ojensi lautasen palvelijalle. — Minä pääsin Lajevskista selville jo tuttavuutemme ensi kuukautena, — jatkoi hän, kääntyen diakonin puoleen. — Me saavuimme tänne yksiin aikoihin. Hänen kaltaisensa ihmiset pitävät hyvin paljon ystävyydestä, läheisestä tuttavuudesta, lujasta yhteisyydentunteesta ja sen semmoisesta, sillä he tarvitsevat aina tovereita kortti-iltoihinsa ja juominkeihinsa; lisäksi he ovat hyvin lavertelevia, ja heillä pitää siis olla kuulijoita. Me tulimme keskenämme ystäviksi, toisin sanoen, hän juoksi luonani joka päivä, esti minua työskentelemästä ja lateli minulle jalkavaimoaan koskevia salaisuuksia. Jo alusta pitäen hämmästyin hänen tavatonta valheellisuuttaan, joka minua suorastaan tympäisi. Ystävänä minä häntä nuhtelin siitä, että hän joi paljon, eli yli varojensa ja teki
- 89. velkaa, että hän kulutti aikaansa toimettomuudessa, ei viitsinyt lueskella, oli epäsivistynyt ja vähätietoinen — ja vastaukseksi kaikkiin moitteisiini hän katkerasti hymyili, huokasi ja lausui: Minä olen kovaosainen, liika ihminen, tahi: Mitäpähän voi vaatia meiltä, maaorjuuden jätteiltä? tahi Me huononemme suvustamme… Taikka hän alkoi ladella pitkää sekasotkua Oneginista, Petshorinista, Byronin Cainista, Dasarovista, joista sanoi: He ovat meidän isiämme lihan ja hengen puolesta. Ymmärrettävä siis muka niin, ettei ole hänen vikansa, jos virkakirjeet ovat viikkomääriä aukaisematta, jos hän itse juo ja juottaa muita, vaan syynä siihen ovat Onegin, Petshorin ja Turgenjev, joka on ensimmäisenä kuvannut nuo kovaosaiset ja liiat ihmiset. Syy hänen tavattomaan irstaisuuteensa ja turmelukseensa ei nähkääs ole hänessä itsessään, vaan jossakin ulkopuolella häntä, avaruudessa. Ja paitsi sitä — sukkela temppu! — irstainen, valheellinen ja viheliäinen ei ole yksistään hän, vaan me… me kahdeksankymmen-luvun ihmiset, me, maaorjuuden veltot, hermostuneet sikiöt, me sivistyksen silpomat… Sanalla sanoen, meidän tulee ymmärtää, että sellainen suuri mies kuin Lajevski on lankeemuksessaankin suuri, että hänen irstaisuutensa, sivistymättömyytensä ja epäsiisteytensä ovat katsottavat välttämättömyyden pyhittämäksi luonnonhistorialliseksi ilmiöksi, että syy siihen on haettava yliluonnollisissa maailman voimissa ja että Lajevskin eteen on ripustettava pyhimyslamppu, koska hän on ajan ja sen tuulahdusten, perinnöllisyyden ynnä muun sellaisen kohtalokas uhri. Kaikki virkamiehet ja naiset, jotka kuuntelivat hänen puheluaan, päästivät senkin seitsemän oh ja ah, mutta minä en hyvään aikaan voinut käsittää, olinko tekemisissä kyynikon vai ovelan veijarin kanssa. Sellaiset ihmiset kuin hän, näöltään älykkäät, jonkun verran kasvatusta saaneet ja omasta jalosukuisuudestaan suurta ääntä pitäväiset, osaavat teeskennellä olevansa syvällisiä luonteita.
- 90. 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