Dna Microarrays For Biomedical Research Methods And Protocols 1st Edition Martin Dufva Phd Msc Auth
- 1. Dna Microarrays For Biomedical Research Methods And Protocols 1st Edition Martin Dufva Phd Msc Auth download https://ebookbell.com/product/dna-microarrays-for-biomedical- research-methods-and-protocols-1st-edition-martin-dufva-phd-msc- auth-2111526 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. Data Analysis Tools For Dna Microarrays Sorin Draghici https://ebookbell.com/product/data-analysis-tools-for-dna-microarrays- sorin-draghici-4342824 Dna Microarrays And Gene Expression Baldi P Hatfield Gw https://ebookbell.com/product/dna-microarrays-and-gene-expression- baldi-p-hatfield-gw-2040756 Dna Microarrays And Related Genomics Techniques Design Analysis And Interpretation Of Experiments Biostatistics 1st Edition David B Allison Editor https://ebookbell.com/product/dna-microarrays-and-related-genomics- techniques-design-analysis-and-interpretation-of-experiments- biostatistics-1st-edition-david-b-allison-editor-2164742 Dna Microarrays Part A Array Platforms And Wetbench Protocols 1st Edition Alan Kimmel https://ebookbell.com/product/dna-microarrays-part-a-array-platforms- and-wetbench-protocols-1st-edition-alan-kimmel-4548510
- 3. Dna Microarrays Gene Expression Applications 1st Edition Bertrand R Jordan Auth https://ebookbell.com/product/dna-microarrays-gene-expression- applications-1st-edition-bertrand-r-jordan-auth-4607918 Dna Microarrays U Nuber https://ebookbell.com/product/dna-microarrays-u-nuber-1019788 Dna Microarrays Databases And Statistics 1st Edition Alan R Kimmel https://ebookbell.com/product/dna-microarrays-databases-and- statistics-1st-edition-alan-r-kimmel-1357522 Dna Microarrays Databases And Statistics Part B Kimmel Ar https://ebookbell.com/product/dna-microarrays-databases-and- statistics-part-b-kimmel-ar-924776 Cancer Diagnostics With Dna Microarrays 1st Edition Steen Knudsen https://ebookbell.com/product/cancer-diagnostics-with-dna- microarrays-1st-edition-steen-knudsen-1473478
- 6. M E T H O D S I N M O L E C U L A R B I O L O G Y TM Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK For other titles published in this series, go to www.springer.com/series/7651
- 7. M E T H O D S I N M O L E C U L A R B I O L O G Y TM DNA Microarrays for Biomedical Research Methods and Protocols Edited by Martin Dufva Department of Micro and Nanotechnology Technical University of Denmark, 2800 Kgs. Lyngby, Denmark
- 8. Editor Martin Dufva Department of Micro and Nanotechnology Technical University of Denmark 2800 Kgs. Lyngby, Denmark Martin.Dufva@nanotech.dtu.dk Series Editor John M. Walker University of Hertfordshire Hatfield, Herts. UK ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-934115-69-5 e-ISBN 978-1-59745-538-1 DOI 10.1007/978-1-59745-538-1 Library of Congress Control Number: 2008938537 # Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer ScienceþBusiness Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper springer.com
- 9. Preface DNA microarray technology has revolutionized research in the past decade. In the beginning microarray technology was mostly used for mRNA expression studies, but soon spread to other applications such as comparative genomic hybridization, SNP and mutation analysis. These applications are now in everyday use in many laboratories and therefore the focus of this volume. It is clear from the protocols in this volume that DNA microarray assays are very complicated to perform even if fabrication of microarray is not considered. It is also clear that there are many different ways to perform microarray assays even if the basic concept is the same, i.e. hybridization of sample DNA (or RNA) to immobilized single stranded capture DNA. Minute changes to a protocol can be pivotal between success and failure in a microarray assays. DNA microarrays fabrication can be divided into two broad categories: on chip synthesis and spotting off chip synthesized DNA. The latter is by far the most common method in house for fabrication of DNA arrays in house. In house fabrication of microarray is necessary when microarrays are not commercially available or is not an economical possibility. The largest providers of microarray are Affymetrix, Illumina and Agilent and all are exemplified in this volume on different kinds of applications. Commercial arrays are typically targeted towards popular organisms and application such as SNP, gene expression analysis, and microarray user that have other requirements are left to fabricate arrays themselves. This volume therefore addresses fabrication issues theoretically as well as giving examples of practical detailed methods. The main advantage of DNA microarray is that many reactions are taking place in parallel on the surface of microarrays. This advantage is also microarray technology’s greatest weakness because all these hybridization reactions need to operate at one single condition applied to the array which put large demands on probe’s choice. Furthermore, we have little knowledge about what is taking place on the surface of microarrays that complicates array development. This volume provides robust protocols for performing microarray assays reproducibly. However, reproducible does not necessarily mean that data obtained correctly reflects what is going on in a cell or an organism. DNA microarray technology is slowly filtering into diagnostic applications that presumably will benefit from miniaturization and highly multiplex assays just like the research community has been doing and will be doing for a considerable time yet. Before microarray comes into clinical use though, we need to find new short and efficient protocols based on the current state-of-the-art protocols provided here. v
- 10. Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix 1. Introduction to Microarray Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Martin Dufva 2. Probe Design for Expression Arrays Using OligoWiz . . . . . . . . . . . . . . . . . . . . . . . 23 Rasmus Wernersson 3. Comparative Genomic Hybridization: Microarray Design and Data Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Richard Redon and Nigel P. Carter 4. Design of Tag SNP Whole Genome Genotyping Arrays. . . . . . . . . . . . . . . . . . . . . 51 Daniel A. Peiffer and Kevin L. Gunderson 5. Fabrication of DNA Microarray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Martin Dufva 6. Immobilization Chemistries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Sascha Todt and Dietmar H. Blohm 7. Fabrication Using Contact Spotter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Annelie Waldén and Peter Nilsson 8. RNA Preparation and Characterization for Gene Expression Studies . . . . . . . . . . . 115 Michael Stangegaard 9. Gene Expression Analysis Using Agilent DNA Microarrays . . . . . . . . . . . . . . . . . . 133 Michael Stangegaard 10. Target Preparation for Genotyping Specific Genes or Gene Segments . . . . . . . . . . 147 Jesper Petersen, Lena Poulsen, and Martin Dufva 11. Genotyping of Mutations in the Beta-Globin Gene Using Allele Specific Hybridization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Lena Poulsen, Jesper Petersen, and Martin Dufva 12. Microarray Temperature Optimization Using Hybridization Kinetics . . . . . . . . . . 171 Steve Blair, Layne Williams, Justin Bishop, and Alexander Chagovetz 13. Whole-Genome Genotyping on Bead Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Kevin L. Gunderson 14. Genotyping Single Nucleotide Polymorphisms by Multiplex Minisequencing Using Tag-Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Lili Milani and Ann-Christine Syvänen 15. Resequencing Arrays for Diagnostics of Respiratory Pathogens . . . . . . . . . . . . . . . 231 Baochuan Lin and Anthony P. Malanoski 16. Comparative Genomic Hybridization: DNA Preparation for Microarray Fabrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Richard Redon, Diane Rigler, and Nigel P. Carter vii
- 11. 17. Comparative Genomic Hybridization: DNA Labeling, Hybridization and Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 Richard Redon, Tomas Fitzgerald, and Nigel P. Carter 18. Chromatin Immunoprecipitation Using Microarrays . . . . . . . . . . . . . . . . . . . . . . . 279 Mickaël Durand-Dubief and Karl Ekwall Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 viii Contents
- 12. Contributors JUSTIN BISHOP University of Utah, Salt Lake City, Utah, USA STEVE BLAIR University of Utah, Salt Lake City, Utah, USA DIETMAR H. BLOHM Department of Biotechnology and Molecular Genetics, Center for Applied Genesensor-Technology (CAG), University of Bremen, Bremen, Germany NIGEL P. CHARTER Wellcome Trust, Sanger Institute, Cambridge, UK ALEXANDER CHAGOVETZ University of Utah, Salt Lake City, Utah, USA MARTIN DUFVA Technical University of Denmark, Kgs. Lyngby, Denmark MICKAËL DURAND-DUBIEF Karolinska Institute /NOVUM, Huddinge, Sweden KARL EKWALL University College Södertörn, Huddinge, Sweden. TOMAS FITZGERALD Wellcome Trust, Sanger Institute, Cambridge, UK KEVIN L. GUNDERSON Illumina, Inc., San Diego, CA, USA BAOCHUAN LIN The Center for Bio/Molecular Science and Engineering, Naval Research Laboratory, Washington, DC, USA ANTHONY P. MALANOSKI The Center for Bio/Molecular Science and Engineering, Naval Research Laboratory, Washington, DC, USA LILI MILANI Uppsala University, Uppsala, Sweden. PETER NILSSON School of Biotechnology, UTH-Royal Institute of Technology, Stockholm, Sweden DANIEL A. PEIFFER Illumina, Inc., San Diego, CA, USA JESPER PETERSEN Technical University of Denmark, Kgs. Lyngby, Denmark LENA POULSEN Technical University of Denmark, Kgs. Lyngby, Denmark RICHARD REDON Wellcome Trust, Sanger Institute, Cambridge, UK DIANE RIGLER Wellcome Trust, Sanger Institute, Cambridge, UK MICHAEL STANGEGAARD University of Copenhagen, Copenhagen, Denmark ANN-CHRISTINE SYVÄNEN University Hospital, Uppsala, Sweden SASHA TODT Center for Applied Genesensor-Technology (CAG), University of Bremen, Bremen, Germany ANNELIE WALDÉN School of Biotechnology, UTH-Royal Institute of Technology, Stockholm, Sweden RASMUS WERNERSSON Center for Biological Sequence Analysis, Technical University of Denmark, Kgs. Lyngby, Denmark LAYNE WILLIAMS University of Utah, Salt Lake City, Utah, USA ix
- 13. Chapter 1 Introduction to Microarray Technology Martin Dufva Abstract DNA microarrays can be used for large number of application where high-throughput is needed. The ability to probe a sample for hundred to million different molecules at once has made DNA microarray one of the fastest growing techniques since its introduction about 15 years ago. Microarray technology can be used for large scale genotyping, gene expression profiling, comparative genomic hybridization and resequencing among other applications. Microarray technology is a complex mixture of numerous technology and research fields such as mechanics, microfabrication, chemistry, DNA behaviour, micro- fluidics, enzymology, optics and bioinformatics. This chapter will give an introduction to each five basic steps in microarray technology that includes fabrication, target preparation, hybridization, detection and data analysis. Basic concepts and nomenclature used in the field of microarray technology and their relationships will also be explained. Key words: Microarray, application, method. 1. Introduction to Microarray Assays Microarray assays originate from traditional solid phase assays––DNA/RNA dot blot assays and Enzyme Linked Immuno Sorbent Assays (ELISA)––that have been used for dec- ades in laboratories. A solid phase assay has molecules attached to a solid support and these molecules are designated ‘capture molecules’ or ‘probes’. The capture molecules are probing the sample for the presence of target molecules. The probe should demonstrate as high specificity and affinity for the target mole- cule as possible. The probes can be PCR products (1), oligonu- cleotides (2), or plasmids or bacterial artificial chromosomes (3) for analysis of genomes and transcriptomes. Though not dis- cussed in this volume, probes can be made of many other types Martin Dufva (ed.), DNA Microarrays for Biomedical Research: Methods and Protocols, vol. 529 ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009 DOI 10.1007/978-1-59745-538-1_1 Springerprotocols.com 1
- 14. of molecules such as proteins (4), antibodies (5, 6), DNA/RNA aptamers (7, 8), small molecules (9) and carbohydrates (10, 11) for analysis of proteomes. The key advantage of microarray technology is that minute amounts of many different probes are immobilized onto a solid support yielding tremendous parallel analysis capacity needed to analyse whole ‘oms’, such as the transcriptome, in one single batch process. Numerous different probes are typically immobi- lized in arrays of spots on a solid support where each spot contains multiple copies of a particular capture molecule/probe. For exam- ple, when fabricating an array it is known that the spot at co- ordinates (x1, y1) contains copies of the probe for gene ‘G’, the target in this case. The identity of probes is therefore encoded by a position in a 2D array (Fig. 1.1). If each spot contains a different probe, a single sample can be probed for the presence of many different target molecules. Typical microarrays contain thousands to million probes on a single ‘chip’ (substrates usually with other dimensions than microscope slide) or microscope slide used for Cell purification 1. Sample preparation Nucleic acids purification Amplification (optional) Labelling Probe choice Surface functionalization Synthesis/spotting Blocking 2. Microarray fabrication 3. Hybridization x1 x2 x3 x4 x5 y1 y2 y3 y4 y5 P1, P2, P3 4. Scanning 5. Data processing Quantification Analysis Biological meaning Fig. 1.1. Layout of the process step for making and using DNA microarray. 2 Dufva
- 15. analysis of the genomes and transcriptomes. Significantly smaller microarrays encompassing 10–100 spots exist as well for sensitive diagnostics of viral and bacterial infections and cost efficient genotyping. The operation whereby sample/target is allowed to react with the probes to generate probe–target interactions is referred to as hybridization. To maximize the sensitivity of the assay, the target should be highly concentrated. Compared to other meth- ods this is not a disadvantage of microarrays because the very small size of microarrays means fairly small quantities of sample/ target are required. The target and the array or probes are then left to react under conditions that facilitate hybridization; typi- cally long hybridization times in high ionic strength buffers at relatively high temperatures. Mixing can be used to increase hybridization kinetics decrease the background binding and obtain homogeneous hybridization over the array. After hybri- dization, arrays are usually subjected to stringent washing pro- cedures to remove cross-hybridizations, i.e. target molecules that have bound to the wrong spot (probes) during the hybridi- zation reaction. Target molecules are labelled using fluorescent molecules or other dyes either pre- or post-hybridization so that probe–target hybridizations can be detected via the generation of a signal. For example, a signal obtained at spot (x1, y1) indicates that gene G (target) is present in the sample. The signal does not provide any other information other than the presence of the target. The size or length of the captured target molecules or the complete sequence/composition of the captured target is not known. This is one weakness of microarray technology as compared to Serial analysis of gene expression (12) and Northern blot analysis. These methods yield target sequence frequencies and target size information, respectively. 2. Early Develop- ment and Origin of Arrays Microarrays offer high-throughput and miniaturized versions of the assay formats they were based on: microtitre plates and dot blot assays. Microtitre plate solid phase assays are based on immo- bilizing specific capture molecules in the wells of a microtitre plate. Each well contains only one type of probe. Thus a well in a microtitre plate assay can be viewed as equivalent to a spot in a microarray. Typically, the probes are immobilized by adding 100–200 mL of probe solution to a well. In comparison, a micro- array spot is typically produced by spotting 0.1–1 nL probe Introduction to Microarray Technology 3
- 16. solution. After removing the excess probe solution, 100–200 mL sample is added to the well. Although microtitre plate assays are highly sensitive, the reaction conditions are not optimal during the assay to obtain maximum sensitivity according to Ekins et al. (13, 14). The reason is the large amount of immobilized probes used in microtitre plate assays capture so many target molecules from the sample that the concentration of the target molecules is decreased. The result is that the density of immobilized target molecules is decreased leading to a lower signal to noise ratio. In microarray assays however, the spots are tiny and contain very small amounts of probe molecules. These small amounts of probe do not affect the concentration of target molecules in the sample. Therefore, microspot assays yield higher density of immo- bilized target molecules and results in higher signal to noise ratios compared to microtitre plate assays. Typically, microarray assays use less than one percent of the target molecules present in the sample (15) . Even though immunoassays were the first to be explored for increased sensitivity and decreased sample requirement (13, 14), microarray-based protein assays did not show promising results until late 1990s (4, 16). DNA arrays started the revolution in the early 1990s where biology transitioned from mainly hypothesis driven research to also include discovery driven research. Micro- arrays are powerful tools for answering questions such as ‘which genes are up-regulated by drug X. . .’. In contrast, classical hypothesis driven research attempted to answer questions such as ‘is drug X regulating Gene Y’. Discovery grade array requires large number of spots per surface area in order to be useful. Arraying DNA onto mem- branes was introduced in 1979 and was referred to as ‘dot blot’ (17). At the time, because of the porosity of membranes and the large spotting volumes, a limitation of the spotting equipment, the density of spots in the arrays was quite low. In other words, only a few different probes were immobilized within a given area of the membrane. Since the spots were at least 1 mm large and the distance between each spot was 1 mm, only 25 different spots could be fit in each cm2 . It was clear that the throughput of dot blot was incapable of extracting the huge amount of information contained within cells. For example, a dot blot assay to deter- mine gene expression of the 40,000 different mRNA in the cell would require a membrane of the size 1600 cm2 corresponding to 160 microscope slides. It is clear that using 160 microscope slides per experiment is cumbersome, expensive and would require 160-fold more samples. Moving from a porous solid support to rigid solid support allowed for the emergence of high density arrays that could be used for discovery driven research (18, 19). 4 Dufva
- 17. 3. Applications 3.1. Gene Expression Arrays Gene expression profiling using DNA microarrays gives informa- tion about the relative differences in gene expression between two different cell populations, e.g. ‘treated’ cells compared to ‘untreated’ cells or cancer cells compared to normal cells. The degree of up- and down-regulation can be estimated for each gene but not the amount (absolute number of molecules) of mRNA expressed in treated cells vs. untreated cells. The first pan-tran- scriptome arrays contained probes towards all the genes in yeast (slightly more than 6,000 transcripts) and were used in several ground breaking publications that demonstrated the usability of DNA microarray technology for genome wide gene expression analysis. Yeast pan-transcriptome arrays were used to determine which genes are involved in the metabolic shift from fermentation to respiration (18), cell cycle (20, 21), sporulation (22) and ploidy (23). After the human genome was sequenced, it was possible to design probes towards known as well as predicted transcripts/ genes in order to get as complete a transcriptome analysis as possible. With arrays capable of analyzing the whole human tran- scriptome, gene expression analysis has been widely used for research on cell physiology and to find diagnostic markers/ mechanisms for diseases such as cancer (24). However, gene expression analysis also has other applications including the deter- mination of water pollution by examining expression profiles in mussels (25), of biocompatibility of surfaces and microchips for cell culture (26, 27) and responses to irradiation (28). Examples of other types of gene expression arrays are exon arrays (29) and siRNA arrays (30, 31) It is not by chance that gene expression analysis was and still is the most used application of microarray technology. Biologi- cally, mRNA levels usually reflect gene function though function of a gene is ultimately determined at the protein activity level. Technically, the transcriptome is easily accessible. Probes made of DNA are easy to obtain and manipulate during fabrication. The transcriptome is relatively well described and limited in size as compared to the corresponding genome. Furthermore, probes can be designed based on expressed sequence tags (EST). The ESTs are generated by sequencing clones of poly A+ molecules in the cell. They can be viewed as a collection of sequences representing the mRNAs that are expressed in a cell at a given time. Conveniently, the whole genome does not need to be sequenced to collect EST data and the first large microarrays were based on ESTs (32, 33). Therefore, transcriptional maps can be generated even for organisms for which there is limited genome sequence data. Introduction to Microarray Technology 5
- 18. 3.2. Genome Wide SNP Analysis and Mutation Analysis of Genes It is estimated that there are about 10 million single nucleotide polymorphisms (SNPs) in the human population. The SNPs are spread throughout the whole genome and can be used as geno- mic markers for finding links between genes and diseases. A SNP is typically a substitution of one base for another specific base. For example, a G is substituted with a C while all other bases in close proximity of the SNP are unchanged. As SNPs can be located inside as well as outside of genes, target needs to be prepared in such a way that the whole genome is represented. This can be major obstacle for such genome wide analysis. Affy- metrix and Illumina provide arrays for genotyping 2.8 million and 1 million SNPs, respectively. The technologies of these companies are based on allele-specific hybridization (34) and allele-specific primer extension, respectively (35). Allele-specific hybridization is based on probes that are centred over the muta- tion site so that the variant base is approximately in the middle of probe. Centering the variant base to the middle of the probe destabilizes mismatch hybrids maximally and the probe will therefore be highly sensitive to mutations in the target. At least two probes are used for detecting a particular SNP: one probe is perfectly matched with one allelic variant and another probe is specific for the other allelic variant. The relative signal strength between the two different probes after stringent hybridization and/or washing is used to assign genotype. Though two probes suffice in principle, Affymetrix uses about 20 probes for each SNP analyzed to obtain enough specificity in the assay (34). Allele-specific primer extension is based on placing the probes so that last nucleotide of the probe is placed over the site of the SNP. A polymerase reaction can be initiated if the probe ends with a perfect match while mismatch hybridization will give a ‘flapping’ 3’ end that cannot serve as an initiation structure for polymerization. The same technologies used for SNP genotyping can be used to genotype mutations that cause monogenetic diseases (36, 37). The drive of array-based assay is to replace automatic sequencing for diagnostics. Depending on the fabrication and detection methods used microarrays can be cheaper, faster and less laborious than automated sequencing. In monogenetic diseases, the gene is mutated in ways that modulate the activity of the corresponding protein. Protein activity can be modu- lated by mutations in the promoter, exon and introns. Typically many mutations can be found near sequences that encode regulatory motifs or catalytic sites of the protein product. The consequence is an increase in the number and complexity of the probes required to genotype a single mutation within such regions as the probe usually overlaps many mutations simultaneously. 6 Dufva
- 19. 3.3. Comparative Genomic Hybridization Comparative genomic hybridizations (CGH) are used to find large deletions and amplifications within genomes (38). CGH was originally based on immobilizing whole chromosomes on glass slides and co-hybridizing different fluorescent labelled con- trols and sample DNAs to the chromosomal preparation. The control DNA originates from cells with normal karyotype while the sample can derive for example from a tumour. The different genetic content in the sample and the control DNA is then resolved using the immobilized condensed chromosomes. The resolution of the original approach is quite low, about 20 Mb in range, due to the use of condensed chromosomes as probes. Array CGH is currently a very popular technique and is based on an immobilized array of probes, much like gene expression arrays. As in the original assay, differently labelled DNAs from sample and control is hybridized to the arrays. In array CGH, the resolution is mainly determined by the number of probes on the array. For example, 32,000 different probes evenly distributed throughout the human genome gives array CGH a resolution of about 0.1 M base (3). CGH arrays can consist of rather large probes produced using BAC clones (3, 39) or smaller cDNA clones (40). Alterna- tively, SNP arrays can be utilized where loss of heterozygocity is taken as proof of a deletion (3, 41). CGH can be used for finding insertion and deletions of chromosomal material (42) or copy number variation analysis (43). 3.4. Array Based Chromatin Immunoprecipitation Assays (ChIP or Chip Assays) Chromatin immunoprecipitation or ChIP Assays are used to find the promoters that bind a specific transcription factor (44) . The principle of ChIP is to crosslink the DNA and proteins together and subsequently isolate DNA fragments that have bound a par- ticular transcription factor using immunoprecipitation with anti- bodies specific to the transcription factor of interest. After amplification, different fragments can be identified on DNA microarrays consisting of probes towards the promoter regions. The ChIP assay is not limited to transcription factors and can also be used for other DNA binding proteins such as histones. 3.5. Other Assays 3.5.1. Sequencing Microarrays can be successfully used for re-sequencing purposes (45, 46). Re-sequencing arrays are in principle the same as SNP array with the exception that four probes are used to determine the bases in a particular site. The variant base is centred in the middle of the probes as described above. There are therefore four probes for each base investigated in a sequence and for example re- sequencing 10 bases in a row requires 40 probes (ten times four). Therefore, sequencing of one million bases requires four million probes. Such re-sequencing can be used for identification of pathogens (45–47) and mutational analysis of mitochondria (48) and genetic variability of genome segments (49). Introduction to Microarray Technology 7
- 20. 3.5.2. Transfection Arrays DNA microarrays can also be used for experiments that are more complex than hybridization reactions. Plasmids contained in gela- tin or other similar matrices can be arrayed by spotting these onto glass microscope slides. Cells are subsequently plated and grown over the surface of the array and a transfection of the plasmids within the array is mediated simultaneously by liposomes (50). Genes involved in apoptosis were efficiently mapped by transfec- tion arrays using 1959 different plasmids spotted on a microscope slide. Subsequent utilization of gene expression arrays on trans- fectants gave information about which proteins were regulators and which were effectors of apoptosis (51). 3.5.3. Template for Protein Array Synthesis Plasmids can also be used as templates for the ‘just in time’ in situ creation of protein microarrays. In such arrays, plasmids carrying different genes, cloned in-frame with the GST gene under the control of a T7 promoter, are spotted together with an antibody towards GST. The arrays of plasmids are transcribed and trans- lated simultaneously using a cell-free lysate such a reticulocyte lysate. The fusion protein is retained on the respective spots by the co-spotted antibody (52). Such arrays can subsequently be used for protein–protein interaction studies. 4. Detailed Description of Microarray Technology Though the principals behind microarray technology seem simple, it is far from easy to perform a ‘complete’ microarray experiment from start to finish. The outline of what is required for a microarray experiment is shown in Fig. 1.1 and discussed in detail below. 4.1. Fabrication Most users are likely not concerned with details concerning the fabrication of DNA microarrays such as probe choice and chem- istry, as in most cases a complete hybridization-ready microarray can be purchased. Manufacturers offering pre-made arrays include Affymetrix, Illumina and Agilent. Microarrays for gene expres- sion, comparative genomic hybridization and detailed SNP ana- lysis are commercially available for a number of popular organisms. However, microarrays for gene expression investiga- tion for the majority of organisms are not commercially available and in these cases, choices regarding probe design, chemistry and fabrication are required. 4.1.1. Probe Choice The production of a DNA microarray that is ready for hybridiza- tion is a complex process. First, probes for microarrays are selected from nucleotide sequence databases such as Genebank. Probe choice strategy is highly dependent on the application and 8 Dufva
- 21. microarray fabrication platform used but a probe should be specific and be able to efficiently capture the target. For gene expression arrays based on 25–60 nucleotide long probes, probe sequences are often chosen from nucleotide sequences found in the 3’end of the transcript. The reason is that cDNA synthesis is often initiated from the 3’ end of the transcript using polyT primers. The results are cDNA fragments that mostly represent the 3’end of the transcripts. In order to maximize signal, the probes are then placed towards the 3’end of the transcript. This results in a bias towards the presence of 3’end sequences. Alter- natively, polyT sequences attached to a solid support can be used to select Poly A transcripts before random primer extension is initiated. If possible gene expression array probes should have the same theoretical melting temperature (Tm) in order to function at the same stringency and similar Gibbs free energy to yield similar hybridization signals between spots (53). cDNA arrays for expression analysis utilize probes that are 1000 nt and the requirement for Tm matching is less of a problem since there is little variation in base composition between 1000 nt sequences. In contrast to probes used for gene expression studies that can be placed ‘somewhere’ in the 3’ end, probes for analysis of muta- tion need to be placed at the site of mutation whether or not it is in a GC rich or a AT rich region. This can put severe restraints on probe selection because it can be difficult to Tm match probes. Most sensitive to this is allele-specific hybridization that requires precise Tm match of probes to discriminate between single base changes. Mutation analysis using allele-specific primer extension or mini-sequencing is not dependent on precise Tm matching but only requires that the probes end either at the nucleotide being investigated or at the nucleotide just before the nucleotide being investigated. Tm matching of probes is easier for SNP analysis than mutation analysis of specific genes. The reason is that SNPs can be chosen with the only criterion that it is a marker for a specific locus. Thus SNP in GC rich and AT rich regions can be avoided. This is not possible to do for mutation analysis of genes since each mutation has been described to have a phenotype and must thus be genotyped. 4.1.2. Immobilization of DNA The solid support and the chemistry used to immobilize probes is very important and may influence the background signal, stability of the bond between the probes and the solid support, probe density, hybridization efficiency, DNA hybrid characteristics, spot morphology, spot density and spot reproducibility (Table 1.1). Typically DNA microarrays are fabricated on a solid glass support because glass is rigid, allows for fluorescent detection as it is trans- parent, and can easily be chemically modified. Polymeric materials have also been considered as solid supports because of the Introduction to Microarray Technology 9
- 22. possibilities to incorporate other approaches, including the use of microfluidic structures within the solid support itself. The flexibility of polymeric materials is a drawback for detection purposes (see below) but the advantage of these is that they are not brittle like glass is. Silicon solid supports can also be utilized for microarray fabrication. There are protocols for making arrays of pre-synthesized oligonucleotides (reviewed in (54) and in Chapter 6) as well as for on-chip synthesis of oligonucleotides on polymeric, glass and silicon solid supports (55–57). Table 1.1 Factors influencing different parameters of a microarray assay Specificity Sensitivity Probe density Morphology Spot density Geometry Microarray fabrication Robotics (X-Y precision) +++ +++ Probe sequence +++ +++ Spotter type (inkjet etc) + + +++ +++ + Spotting conditions ++ ++ +++ +++ + Immobilization chemistry ++ +++ +++ +++ ++ ++ Probe conc + ++ +++ + + Spotting buffer + + ++ ++ ++ + Target prep Cell purity +++ +++ Nucl. Acids purification ++ +++ Amplification + +++ + Labelling + +++ Hyb cond. Mixing + +++ + Stringency + ++ S/N Background + +++ Detection methods ++ +++ 10 Dufva
- 23. 4.2. Target Preparation Target preparation is a complex procedure that is always the responsibility of the end user of the microarray. Inappropriate target preparation limits the potential of the microarray experi- ment even if high quality microarrays and advanced bioinformatic systems are used. Target preparation is usually a multi-step process that can be divided into cell and nucleic acids purification, ampli- fication and labelling (Fig. 1.1). The number of steps required is dependent on the application and the biological material at hand. 4.2.1. Cell Preparation Cells can be selected from complex matrixes such as tissue or blood using laser microdissection (58) or antibody affinity purifi- cation, respectively. For gene expression applications, the selec- tion of target cells is very important because analyzing a mixture of different cells will result in an average gene expression profile of the mixture(normal and treated cells) and important gene regula- tions can be missed. CGH analysis may also require selection of cells. Analysis of complex tissues such as tumours requires purifi- cation of cancer cells from the surrounding healthy tissue prior to analysis. Normal diploid cells will reduce the amplitude of the signal coming from amplifications and/or deletions of the tumour cells. Analysis of inherited chromosomal changes is by contrast not dependent at all on the type of diploid cell analyzed and requires no purification. 4.2.2. Nucleic Acid Purification The aim of nucleic acid purification is to prepare sufficient amount of either DNA or RNA to levels of such purity that it can be used in enzymatic reactions. DNA is the least sensitive nucleic acid and can readily be prepared from fresh or frozen tissue materials in sufficient amounts using a number of different methods. Archived material such as paraffin embedded tissue slices can also be used but yield DNA of lesser quality (59). By contrast, RNA is more sensitive as it is easily degraded by endogenous nucleases (RNases). RNA preparation methods must inhibit RNase activity. Often, guanidinium isothiocyanate (GITC)-based methods are used in the RNA preparation protocols to avoid RNase activity (60). Though RNase activity is inhibited during the preparation protocol, RNA may be degraded after the sample is taken if it is not snap frozen or directly lysed in GITC containing lysis buffer. In this buffer, RNA is stable for long-term storage at 20 C (61). 4.2.3. Amplification Often there is a need to assay from limited amounts of sample material. Large arrays for the investigation of genome wide SNP analysis, CGH or gene expression analysis typically require large amounts of target in order to give sufficient hybridization signal. As such, there is a need to amplify the DNA or RNA prior to labelling of the target. The genome can first be cut into fragments with restriction enzymes and primer sequences subsequently ligated to the frag- ments. The complex DNA can then be amplified using PCR. This Introduction to Microarray Technology 11
- 24. amplification method requires little starting material, 250 ng, but reduces the complexity of the resulting target (34). Alternatively, Phi29- based random primed isothermal wide genome amplifica- tion is a non-PCR-based method that gives better coverage of complex genome and allows genome wide genetic analysis from 100 ng DNA samples (35). mRNA can also be amplified but only indirectly. The most popular method is to reverse transcribe mRNA into cDNA using polyT primers modified with T7 promoter sequences in the 5’end. Double stranded DNA is then generated where each fragment carries a T7 viral promoter that can be utilized for T7 in vitro transcription reaction (IVT). An 80-fold amplification of the mRNA can be achieved using this method and the amplified product is designated ‘aRNA’ (62). The method has since been modified to use two rounds of amplification resulting in an approximately millionfold amplification of the target (63). This is sufficient for generating gene expression profiles from single cells (64). Besides the large amplification achieved by this method, another attractive feature is that it generates single stranded tar- gets. This in part explains the efficiency of this amplification method for microarray analysis. Even though the target is gener- ated using random priming and is by nature a linear amplification form, it usually represents the original mRNA population qualita- tively and quantitatively less well than cDNA directly reverse transcribed (65). The reason is that there is always a selection during enzymatic reactions and the selection is enhanced by the amplification process. It is therefore not unexpected that the correlation between the gene expression profile obtained using different target preparation methods is gradually decreased with increasing amplification of the target (65). Though reverse tran- scription is the gold standard for target preparation for microarray experiments, it has been shown that reverse transcription can also introduce systematic biases in array experiment as compared to hybridization of labelled mRNA (66). 4.2.4. Labelling Typically, target hybridized to an array is labelled with fluorescent molecules or biotins for post-hybridization staining. For gene expression analysis, the target is usually labelled during cDNA synthesis from RNA by spiking labelled nucleotides into the reverse transcription reaction. Similarly, aRNA ready for hybridi- zation contains labelled ribonucleotides that are incorporated during the IVT reaction. Alternative approaches have been used to avoid enzymatic treatment of mRNA prior to hybridization. These involve direct labelling of RNA prior to hybridization (66) or to stain the RNA post-hybridization with gold nanoparticles covered with poly-T oligonucleotides (68). 12 Dufva
- 25. There are similar ways to label DNA prior to hybridization for genetic tests. Direct labelling of PCR primers is a rapid and con- venient method to introduce a label during the amplification process. Alternatively, labelled nucleotides can be spiked into a labelling reaction to give random labelling of the fragments. The latter method typically introduces several labels per strands com- pared to end labelled primers. 4.3. Hybridization Reactions Microarrays were first hybridized under cover slips using 2 mL of highly concentrated target solution per cm2 (1). This hybridiza- tion method relies solely on diffusion of target molecules to the corresponding spot. Since target molecules are fairly large, the diffusion time from one side of the array to the other takes many years, it can be expected that the spot only reacts with targets that are present in a fraction of the total hybridization volume. Hybri- dization without mixing often results in heterogeneous reaction conditions over the array surface. It would therefore be advanta- geous to perform mixing on arrays. A literary survey indicates several solutions including; cavitation micro streaming (69), mag- netic bar stirring (70), air driven bladders (15), centrifugal mixing (71) and shear driven mixing (72). A drawback of most mixing strategies is that the sample is significantly diluted in order to be mixed using the above described methods compared to static hybridization using very small volumes. Despite dilution, mixing often gives a 2–10-fold increase in signal, provides homogeneous hybridization conditions over the entire array and lowers back- ground signals. 4.4. Detection By far the most popular method for detection of hybridized array is fluorescence. Fluorochromes offer high sensitivity, large dynamic range, are easy to work with and a single array can be stained with up to four different fluorochromes, each with distinct spectral properties. The drawbacks of fluorescence are photo bleaching during exposure and decomposition of fluorochromes over time. These make fluorescent stains less suitable for long-term archiving of hybridized slides. Since a microarray assay is generally not quan- titative, each sample must be compared with a control. Gene expression arrays can be used with either one or two fluorescent dyes. The use of a single fluorescent dye for detection requires that the sample and the control are hybridized on two different slides, whereas the use of two different fluorescent dyes allows the use of a single slide to probe both the sample and control targets. Utilization of three fluorochromes gives opportunities for quality control. One of the dyes is only used as an indicator of the presence and relative quantity of immobilized probes while the other two fluorochromes can be used for sample and control target labelling and detection (73, 74). Missing spot or poor quality spots can then be easily filtered out prior to analysis. Introduction to Microarray Technology 13
- 26. Four different fluorochromes might be used to detect the four different nucleotides that can be incorporated in DNA as in mini- sequencing reactions (75). Although fluorescence is popular, the method requires fairly expensive scanners and many applications such as gene expression profiling and CGH also need better assay detection limits. There- fore many other approaches for microarray detection have been proposed. Light scattering of silver enhanced gold nanoparticles has several orders of magnitude better detection limits compared to traditional fluorescent detection and allows for SNP analysis from unamplified genomic DNA (76) and gene expression analysis from as little as 0.5 mg of unamplified total RNA (68). Hesse et al. have recently demonstrated a novel but not yet commercially available fluorescent-based scanner system that gives similar or better sensi- tivity to scattering of light by gold/silver particles for the detection of single molecules hybridized to microarray spots (77). Measuring absorbance provides inexpensive solution for detection of analyte binding to DNA microarray. Suitable stain- ing methods include gold nano particles (78), gold/silver parti- cles (79) and enzyme based stains such as alkaline phosphatase BCIP/NBT reactions (80, 81). Such stains are visible provided that the spot is sufficiently large but can conveniently be digitized with an inexpensive 1200 dpi or better flatbed scanner. The draw- backs of these staining methods are significantly higher detection limit and limited dynamic range of the assays. 4.5. Data Analysis 4.5.1. Quantification After digitalization using a scanner, the images (usually TIF files) are analyzed in specialized software. The relative fluorescence of a spot is quantified by calculating the ‘whiteness’ of the spot. This is simply done by calculating the pixel values within a spot where black is equal to no signal, white is maximum signal and the different gray scales correspond to everything in between minimum and maximum signal. Defining a spot and the pixels that should be counted is not easy. Some software requires that each spot is defined by the user by encircling the spot (the freeware Scanalyze (82) is an example). Once the spot is defined, the pixel value of every pixel within the spot is summed to give a total spot signal or summed and divided by the number of pixels in order to give the density of the fluorescence. Spots of different sizes constitute a problem and usually require user intervention, a cumbersome and time intensive process. Spotfinder, another freeware, only allows pixels above the background signal level to be included in the spot. In Spotfinder, spots with severely malformed morphologies such as ‘coffespots’ and ‘halfmoon’ shapes can be still be quantified. The background signal originating from dark currents of the instrument, substrate chemistries as well as unspecific binding of target to the substrate surface can be calculated in different ways. 14 Dufva
- 27. Local background signal is calculated by examining pixels sur- rounding each spot. The fluorescent value of the spot is then calculated to be (signal from spot – signal from the background). This is a very good method to use if the microarray slide has uneven background. A simpler approach is to subtract the average background value of the entire microarray slide from each of the spots on the array. 4.5.2. Normalization A drawback of co-hybridization in microarray-based CGH and gene expression assays is that different fluorescent dyes are not perfectly matched in terms of quantum yields and sensitivity to light and ozone. For instance, there is a non-linear relationship between the signals from Cy3 and Cy5, the most popular dyes for detection on arrays. These differences must be compensated for by normalization software such as QSPLINE (83). Using the same dye for both the sample and control does not have the above problem. In this case, differences between hybridization reactions must be compensated for in silico. After normalization, the data set can be analyzed by various statistical methods. 5. Parameters Used to Describe Microarray Assays 5.1. Geometry Geometry of the array refers to how well the spots are ordered into an array; i.e. how even/equal are the distances between each spot. Even spacing between the spots is important because misaligned spots are difficult to quantify automatically. Geometry of the array is mainly determined by the precision of the machinery used to fabricate the arrays. For spotted arrays, geometry can be affected when spotting on hydrophobic surfaces because the droplet can ‘move’ from the point where it was deposited. 5.2. Spot Density Spot density is defined as the number of spots per unit area. Spot density is determined by the precision of the machinery to localize a spot in (x,y) co-ordinate system, probe solution ejection system, immobilization chemistry and the composition of spotting solu- tion (probe concentration and the spotting buffer). In most cases it is the size of the spots that determines spot density. Spot size can be altered by changing the spotting volume. Spot volumes are determined by the deposition technique used to deliver the dro- plets to the surface, the hydrophobicity of the surface (determined by the chemistry) and the composition of the spotting buffer. For instance, spots on hydrophobic surfaces can be made larger by adding appropriate amount of detergents in the spotting buffer. Introduction to Microarray Technology 15
- 28. Increased spot density is required in recent years to meet the demands of the ever increasing complexity of microarray experi- ments. The present drivers for increasing spot density are genome and proteome analysis and not gene expression arrays. Chapter 5 discusses methods to fabricate arrays with higher spot density/ more compact arrays. 5.3. Probe Density Probe density is defined as the number of probe molecules per unit area. It is a measurement used to characterize immobilization chemistries that link DNA to microarray surfaces. The probe density is determined by the chemistry of the surface, modifica- tions made to the DNA to increase immobilization, the size of the molecules to be immobilized, the spotting buffer and the probe concentration (Table 1.1). The chemistry of the surface and the probes determines the efficiency of immobilization of the probes. Spotting buffers also need to be chosen correctly so that the spotting buffer does not interfere with the chemistry. For instance, TRIS buffers need to be avoided if the surface contains epoxy or aldehyde funtionalization because TRIS contains amines that can react with the surface before the DNA has a chance to bind. The spotted probe con- centration is very important and too low a concentration will give spots with few capture molecules and thus low maximum signal is usually translated to low sensitivity. Optimal spotted probe con- centration often needs to be titrated for each surface chemistry. Critical to high probe density is the deposition of the correct drop volume on the surface (see also Section 5.2) so that the spots, when dried, contain at least a monolayer of molecules that can be attached to the surface. Probe density is usually optimized to maximize hybridization signal but not hybridization efficiency. This is to produce spots with as large a dynamic range and as high a sensitivity as possible. Probe density also determines the upper limit of the possible hybridizations within a spot. In many cases, only a fraction of the probes immobilized to the surface can undergo hybridization even under saturated conditions. The den- sity of hybridized targets to a surface is referred to as ‘hybridized density’. 5.4. Sensitivity The sensitivity of a microarray assay is defined as the lowest con- centration of target molecules that can be detected on a spot. Sensitivity is affected by all factors of a microarray experiment and is the parameter that most users have problems with. In particular, users that set up their own microarray assays have to consider all factors that could influence sensitivity (Table 1.1). Users who buy ready made arrays are limited to optimize and/or select appropriate target preparation methods, detection systems and/or hybridization conditions to increase the sensitivity. 16 Dufva
- 29. Probe density and organization on the surface as well as the affinity of the probes is determined during the fabrication phase. Probe affinity should be high. This is predicted using the calculated G values of the probes (84). The probes must also be immobi- lized in the correct density to obtain maximum signal. Factors affecting probe density are discussed above (Section 5.3). Low signal on arrays can result from suboptimal probe function, which can be caused by the molecular organization of the probes on the surface. It is well known that the use of molecular spacers, to move short probes away from the substrate surface gives better signal than short probes directly linked to surfaces. Target preparation is very critical to produce highly sensitive assays. As previously discussed, the target preparation method used determines the concentration of target molecules to be hybridized to the array which in turn determines the sensitivity of the assay. Mixing benefits sensitivity because it moves target molecules from one end of the microarray to the other; something not possible by diffusion alone. Finally, the detection system and method has a large impact on sensitivity (see Section 4.4). Usually, instrumentation is fixed because of the high cost associated with acquiring new equip- ment. However, the sensitivity of the detector in the instruments can, in most cases, be adjusted to appropriate levels to obtain the highest assay sensitivity. For example, for weakly fluorescent arrays the sensitivity of instruments can easily be adjusted so that the array source is exposed to more excitation light resulting in more light emission from the arrays. However, it should be noted that the background signal usually increases as well when the sensitivity of the instrument is increased. 5.5. Specificity Specificity (or selectivity) is defined as how selective the probes are to capture the intended target in a complex background of other target molecules. Targets with similar but not identical sequences may bind to the probe intended for the intended target. The binding of an ‘unintended’ target to a probe is called cross-hybri- dization. Cross-hybridization must be minimized as it decreases the diagnostic power of genotyping arrays and the resolution of up-and down-regulation of genes in gene expression microarray experiments. In traditional Northern blot assays specificity is obtained by the sequence and length of the target. Thus unspecific hybridization to other fragments than the intended one is easily observed because these fragments typically are of different lengths. The specificity of microarrays is based on the sequence of the probes, therefore probe selection is critical. Closely related to probe selection is the stringency level of the assay. Probes should be chosen so that all the probes on the array function optimally at a single stringency condition, i.e. on determined buffer concentration and temperature. Optimal condition for Introduction to Microarray Technology 17
- 30. probe function can be predicted by calculating the melting tem- perature of the probes. Choosing appropriate stringency is usually a balance between having high signals on the arrays and sufficient specificity. Approaches other than the probe sequence can be used to maximize the specificity of the assay. One is to remove unwanted nucleic acids from a complex target. This can be achieved for some applications by selecting only cells of interest using immuno capture and thereby removing the ‘contaminating’ cells from the assay. For monogenetic diseases or viral/bacterial diagnostics, it is very convenient to ‘select’ nucleic acids to be analyzed using PCR. Acknowledgements Thanks are due to David Sabourin, Lena Poulsen and Jesper Petersen for reading the manuscript and helpful discussions. References 1. Schena, M., Shalon, D., Davis, R.W. and Brown, P.O. (1995) Quantitative monitor- ing of gene expression patterns with a com- plementary DNA microarray. Science, 270, 467–470. 2. Zammatteo, N., Jeanmart, L., Hamels, S., Courtois, S., Louette, P., Hevesi, L. and Remacle, J. (2000) Comparison between dif- ferent strategies of covalent attachment of DNA to glass surfaces to build DNA micro- arrays. Anal Biochem, 280, 143–150. 3. Ishkanian, A.S., Malloff, C.A., Watson, S.K., DeLeeuw, R.J., Chi, B., Coe, B.P., Snijders, A., Albertson, D.G., Pinkel, D., Marra, M.A. et al. (2004) A tiling resolution DNA micro- array with complete coverage of the human genome. Nat Genet, 36, 299–303. 4. Lueking, A., Horn, M., Eickhoff, H., Bus- sow, K., Lehrach, H. and Walter, G. (1999) Protein microarrays for gene expression and antibody screening. Anal Biochem, 270, 103–111. 5. Huang, R.P. (2001) Detection of multiple proteins in an antibody-based protein micro- array system. J Immunol Methods, 255, 1–13. 6. Huang, R.P., Huang, R., Fan, Y. and Lin, Y. (2001) Simultaneous detection of multiple cytokines from conditioned media and patient’s sera by an antibody-based protein array system. Anal Biochem, 294, 55–62. 7. Li, Y., Lee, H.J. and Corn, R.M. (2006) Fabrication and characterization of RNA aptamer microarrays for the study of pro- tein-aptamer interactions with SPR imaging. Nucleic Acids Res, 34, 6416–6424. 8. Lee, M. and Walt, D.R. (2000) A fiber-optic microarray biosensor using aptamers as receptors. Anal Biochem, 282, 142–146. 9. Bradner, J.E., McPherson, O.M., Mazitschek, R., Barnes-Seeman, D., Shen, J.P., Dhaliwal, J., Stevenson, K.E., Duffner, J.L., Park, S.B., Neuberg, D.S. et al. (2006) A robust small- molecule microarray platform for screening cell lysates. Chem Biol, 13, 493–504. 10. Wang, D., Liu, S., Trummer, B.J., Deng, C. and Wang, A. (2002) Carbohydrate micro- arrays for the recognition of cross-reactive molecular markers of microbes and host cells. Nat Biotechnol, 20, 275–281. 11. Horlacher, T. and Seeberger, P.H. (2006) The utility of carbohydrate microarrays in glycomics. Omics, 10, 490–498. 12. Velculescu, V.E., Zhang, L., Vogelstein, B. and Kinzler, K.W. (1995) Serial analysis of gene expression. Science, 270, 484–487. 13. Ekins, R., Chu, F. and Biggart, E. (1990) Fluorescence spectroscopy and its applica- tion to a new generation of high sensitivity, multi-microspot, multianalyte, immunoas- say. Clin Chim Acta, 194, 91–114. 18 Dufva
- 31. 14. Ekins, R., Chu, F. and Biggart, E. (1990) Multispot, multianalyte, immunoassay. Ann Biol Clin (Paris), 48, 655–666. 15. Adey, N.B., Lei, M., Howard, M.T., Jensen, J.D., Mayo, D.A., Butel, D.L., Coffin, S.C., Moyer, T.C., Slade, D.E., Spute, M.K. et al. (2002) Gains in sensitivity with a device that mixes microarray hybridization solution in a 25-microm-thick chamber. Anal Chem, 74, 6413–6417. 16. MacBeath, G. and Schreiber, S.L. (2000) Printing proteins as microarrays for high- throughput function determination. Science, 289, 1760–1763. 17. Kafatos, F.C., Jones, C.W. and Efstratiadis, A. (1979) Determination of nucleic acid sequence homologies and relative concentra- tions by a dot hybridization procedure. Nucleic Acids Res, 7, 1541–1552. 18. DeRisi, J.L., Iyer, V.R. and Brown, P.O. (1997) Exploring the metabolic and genetic control of gene expression on a genomic scale. Science, 278, 680–686. 19. Iyer, V.R., Eisen, M.B., Ross, D.T., Schuler, G., Moore, T., Lee, J.C., Trent, J.M., Staudt, L.M., Hudson, J., Jr., Boguski, M.S. et al. (1999) The transcriptional pro- gram in the response of human fibroblasts to serum. Science, 283, 83–87. 20. Cho, R.J., Campbell, M.J., Winzeler, E.A., Steinmetz, L., Conway, A., Wodicka, L., Wolfsberg, T.G., Gabrielian, A.E., Lands- man, D., Lockhart, D.J. et al. (1998) A gen- ome-wide transcriptional analysis of the mitotic cell cycle. Mol Cell, 2, 65–73. 21. Spellman, P.T., Sherlock, G., Zhang, M.Q., Iyer, V.R., Anders, K., Eisen, M.B., Brown, P.O., Botstein, D. and Futcher, B. (1998) Comprehensive identification of cell cycle- regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. Mol Biol Cell, 9, 3273–3297. 22. Chu, S., DeRisi, J., Eisen, M., Mulholland, J., Botstein, D., Brown, P.O. and Herskowitz, I. (1998) The transcriptional program of sporula- tion in budding yeast. Science, 282, 699–705. 23. Galitski, T., Saldanha, A.J., Styles, C.A., Lander, E.S. and Fink, G.R. (1999) Ploidy regulation of gene expression. Science, 285, 251–254. 24. Perez-Diez, A., Morgun, A. and Shulzhenko, N. (2007) Microarrays for cancer diagnosis and classification. Advances in experimental medicine and biology, 593, 74–85. 25. Venier, P., De Pitta, C., Pallavicini, A., Mar- sano, F., Varotto, L., Romualdi, C., Don- dero, F., Viarengo, A. and Lanfranchi, G. (2006) Development of mussel mRNA pro- filing: Can gene expression trends reveal coastal water pollution? Mutat Res. 26. Stangegaard, M., Wang, Z., Kutter, J.P., Dufva, M. and Wolff, A. (2006) Whole gen- ome expression profiling using DNA micro- array for determining biocompatibility of polymeric surfaces. Molecular Biosystems, 2, 421–428. 27. Stangegaard, M., Petronis, S., Jorgensen, A.M., Christensen, C.B. and Dufva, M. (2006) A biocompatible micro cell culture chamber (microCCC) for the culturing and on-line monitoring of eukaryote cells. Lab Chip, 6, 1045–1051. 28. Tusher, V.G., Tibshirani, R. and Chu, G. (2001) Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A, 98, 5116–5121. 29. Frey, B.J., Mohammad, N., Morris, Q.D., Zhang, W., Robinson, M.D., Mnaimneh, S., Chang, R., Pan, Q., Sat, E., Rossant, J. et al. (2005) Genome-wide analysis of mouse transcripts using exon microarrays and factor graphs. Nat Genet, 37, 991–996. 30. Castoldi, M., Schmidt, S., Benes, V., Noer- holm, M., Kulozik, A.E., Hentze, M.W. and Muckenthaler, M.U. (2006) A sensitive array for microRNA expression profiling (miChip) based on locked nucleic acids (LNA). Rna, 12, 913–920. 31. Beuvink, I., Kolb, F.A., Budach, W., Gar- nier, A., Lange, J., Natt, F., Dengler, U., Hall, J., Filipowicz, W. and Weiler, J. (2007) A novel microarray approach reveals new tissue-specific signatures of known and predicted mammalian microRNAs. Nucleic Acids Res. 32. Schena, M., Shalon, D., Heller, R., Chai, A., Brown, P.O. and Davis, R.W. (1996) Paral- lel human genome analysis: microarray- based expression monitoring of 1000 genes. Proc Natl Acad Sci U S A, 93, 10614–10619. 33. Loftus, S.K., Chen, Y., Gooden, G., Ryan, J.F., Birznieks, G., Hilliard, M., Baxevanis, A.D., Bittner, M., Meltzer, P., Trent, J. et al. (1999) Informatic selection of a neural crest-melanocyte cDNA set for microarray analysis. Proc Natl Acad Sci U S A, 96, 9277–9280. Introduction to Microarray Technology 19
- 32. 34. Matsuzaki, H., Dong, S., Loi, H., Di, X., Liu, G., Hubbell, E., Law, J., Berntsen, T., Chadha, M., Hui, H. et al. (2004) Genotyp- ing over 100,000 SNPs on a pair of oligonu- cleotide arrays. Nat Methods, 1, 109–111. 35. Gunderson, K.L., Steemers, F.J., Lee, G., Mendoza, L.G. and Chee, M.S. (2005) A genome-wide scalable SNP genotyping assay using microarray technology. Nat Genet, 37, 549–554. 36. Chou, W.H., Yan, F.X., Robbins-Weilert, D.K., Ryder, T.B., Liu, W.W., Perbost, C., Fairchild, M., de Leon, J., Koch, W.H. and Wedlund, P.J. (2003) Comparison of two CYP2D6 genotyping methods and assess- ment of genotype-phenotype relationships. Clin Chem, 49, 542–551. 37. Gemignani, F., Perra, C., Landi, S., Canzian, F., Kurg, A., Tonisson, N., Galanello, R., Cao, A., Metspalu, A. and Romeo, G. (2002) Reliable detection of beta-thalasse- mia and G6PD mutations by a DNA micro- array. Clin Chem, 48, 2051–2054. 38. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F. and Pinkel, D. (1992) Comparative genomic hybridization for molecular cytogenetic ana- lysis of solid tumors. Science, 258, 818–821. 39. Fiegler, H., Redon, R., Andrews, D., Scott, C., Andrews, R., Carder, C., Clark, R., Dovey, O., Ellis, P., Feuk, L. et al. (2006) Accurate and reliable high-throughput detec- tion of copy number variation in the human genome. Genome Res, 16, 1566–1574. 40. Pollack, J.R., Perou, C.M., Alizadeh, A.A., Eisen, M.B., Pergamenschikov, A., Williams, C.F., Jeffrey, S.S., Botstein, D. and Brown, P.O. (1999) Genome-wide analysis of DNA copy-number changes using cDNA microar- rays. Nat Genet, 23, 41–46. 41. Barrett, M.T., Scheffer, A., Ben-Dor, A., Sampas, N., Lipson, D., Kincaid, R., Tsang, P., Curry, B., Baird, K., Meltzer, P.S. et al. (2004) Comparative genomic hybridization using oligonucleotide microarrays and total genomic DNA. Proc Natl Acad Sci U S A, 101, 17765–17770. 42. Pinkel, D. and Albertson, D.G. (2005) Array comparative genomic hybridization and its applications in cancer. Nat Genet, 37 Suppl, S11–17. 43. Redon, R., Ishikawa, S., Fitch, K.R., Feuk, L., Perry, G.H., Andrews, T.D., Fiegler, H., Shapero, M.H., Carson, A.R., Chen, W. et al. (2006) Global variation in copy num- ber in the human genome. Nature, 444, 444–454. 44. Ren, B., Robert, F., Wyrick, J.J., Aparicio, O., Jennings, E.G., Simon, I., Zeitlinger, J., Schreiber, J., Hannett, N., Kanin, E. et al. (2000) Genome-wide location and function of DNA binding proteins. Science, 290, 2306–2309. 45. Malanoski, A.P., Lin, B., Wang, Z., Schnur, J.M. and Stenger, D.A. (2006) Automated identification of multiple micro-organisms from resequencing DNA microarrays. Nucleic Acids Res, 34, 5300–5311. 46. Lin, B., Wang, Z., Vora, G.J., Thornton, J.A., Schnur, J.M., Thach, D.C., Blaney, K.M., Ligler, A.G., Malanoski, A.P., San- tiago, J. et al. (2006) Broad-spectrum respiratory tract pathogen identification using resequencing DNA microarrays. Gen- ome Res, 16, 527–535. 47. Lin, B., Blaney, K.M., Malanoski, A.P., Lig- ler, A.G., Schnur, J.M., Metzgar, D., Russell, K.L. and Stenger, D.A. (2007) Using a rese- quencing microarray as a multiple respiratory pathogen detection assay. J Clin Microbiol, 45, 443–452. 48. Chee, M., Yang, R., Hubbell, E., Berno, A., Huang, X.C., Stern, D., Winkler, J., Lock- hart, D.J., Morris, M.S. and Fodor, S.P. (1996) Accessing genetic information with high-density DNA arrays. Science, 274, 610–614. 49. Wang, D.G., Fan, J.B., Siao, C.J., Berno, A., Young, P., Sapolsky, R., Ghandour, G., Per- kins, N., Winchester, E., Spencer, J. et al. (1998) Large-scale identification, mapping, and genotyping of single-nucleotide poly- morphisms in the human genome. Science, 280, 1077–1082. 50. Ziauddin, J. and Sabatini, D.M. (2001) Microarrays of cells expressing defined cDNAs. Nature, 411, 107–110. 51. Palmer, E.L., Miller, A.D. and Freeman, T.C. (2006) Identification and characterisa- tion of human apoptosis inducing proteins using cell-based transfection microarrays and expression analysis. BMC Genomics, 7, 145. 52. Ramachandran, N., Hainsworth, E., Bhullar, B., Eisenstein, S., Rosen, B., Lau, A.Y., Wal- ter, J.C. and LaBaer, J. (2004) Self-assem- bling protein microarrays. Science, 305, 86–90. 20 Dufva
- 33. 53. Bruun, G.M., Wernersson, R., Juncker, A.S., Willenbrock, H. and Nielsen, H.B. (2007) Improving comparability between microar- ray probe signals by thermodynamic inten- sity correction. Nucleic Acids Res. 54. Dufva, M. (2005) Fabrication of high quality microarrays. Biomol Eng, 22, 173–184. 55. Moorcroft, M.J., Meuleman, W.R., Latham, S.G., Nicholls, T.J., Egeland, R.D. and Southern, E.M. (2005) In situ oligonucleo- tide synthesis on poly(dimethylsiloxane): a flexible substrate for microarray fabrication. Nucleic Acids Res, 33, e75. 56. Southern, E.M., Case-Green, S.C., Elder, J.K., Johnson, M., Mir, K.U., Wang, L. and Williams, J.C. (1994) Arrays of complemen- tary oligonucleotides for analysing the hybri- disation behaviour of nucleic acids. Nucleic Acids Res, 22, 1368–1373. 57. Fodor, S.P., Read, J.L., Pirrung, M.C., Stryer, L., Lu, A.T. and Solas, D. (1991) Light-directed, spatially addressable parallel chemical synthesis. Science, 251, 767–773. 58. Ohyama, H., Zhang, X., Kohno, Y., Alevi- zos, I., Posner, M., Wong, D.T. and Todd, R. (2000) Laser capture microdissection- generated target sample for high-density oligonucleotide array hybridization. Biotech- niques, 29, 530–536. 59. Gilbert, M.T., Haselkorn, T., Bunce, M., Sanchez, J.J., Lucas, S.B., Jewell, L.D., Van Marck, E. and Worobey, M. (2007) The isolation of nucleic acids from fixed, paraf- fin-embedded tissues-which methods are useful when? PLoS ONE, 2, e537. 60. Chomczynski, P. and Sacchi, N. (1987) Sin- gle-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloro- form extraction. Anal Biochem, 162, 156–159. 61. Gilleland, R.C. and Hockett, R.D., Jr. (1998) Stability of RNA molecules stored in GITC. Biotechniques, 25, 944–946, 948. 62. Van Gelder, R.N., von Zastrow, M.E., Yool, A., Dement, W.C., Barchas, J.D. and Eber- wine, J.H. (1990) Amplified RNA synthe- sized from limited quantities of heterogeneous cDNA. Proc Natl Acad Sci U S A, 87, 1663–1667. 63. Eberwine, J., Yeh, H., Miyashiro, K., Cao, Y., Nair, S., Finnell, R., Zettel, M. and Cole- man, P. (1992) Analysis of gene expression in single live neurons. Proc Natl Acad Sci U S A, 89, 3010–3014. 64. Kamme, F., Zhu, J., Luo, L., Yu, J., Tran, D.T., Meurers, B., Bittner, A., Westlund, K., Carlton, S. and Wan, J. (2004) Single-cell laser-capture microdissection and RNA amplification. Methods Mol Med, 99, 215–223. 65. Hughes, T.R., Mao, M., Jones, A.R., Burch- ard, J., Marton, M.J., Shannon, K.W., Lef- kowitz, S.M., Ziman, M., Schelter, J.M., Meyer, M.R. et al. (2001) Expression profil- ing using microarrays fabricated by an ink-jet oligonucleotide synthesizer. Nat Biotechnol, 19, 342–347. 66. Gupta, V., Cherkassky, A., Chatis, P., Joseph, R., Johnson, A.L., Broadbent, J., Erickson, T. and DiMeo, J. (2003) Directly labeled mRNA produces highly precise and unbiased differential gene expression data. Nucleic Acids Res, 31, e13. 67. Stangegaard, M., Dufva, I.H. and Dufva, M. (2006) Reverse transcription using random pentadecamer primers increases yield and quality of resulting cDNA. Biotechniques, 40, 649–657. 68. Huber, M., Wei, T.F., Muller, U.R., Lefebvre, P.A., Marla, S.S. and Bao, Y.P. (2004) Gold nanoparticle probe-based gene expression analysis with unamplified total human RNA. Nucleic Acids Res, 32, e137. 69. Liu, R., Lenigk, R., Druyor-Sanchez, R., Yang, J. and Grodzinski, P. (2003) Hybridi- zation enhancement using cavitation micro- streaming. Anal Chem, 75, 1911–1917. 70. Yuen, P., Li, G., Bao, Y. and Muller, U. (2003) Microfluidic devices for fluidic circulation and mixing improve hybridization signal intensity on DNA arrays. Lab on a Chip, 3, 46–50. 71. Bynum, M.A. and Gordon, G.B. (2004) Hybridization enhancement using microflui- dic planetary centrifugal mixing. Anal Chem, 76, 7039–7044. 72. Pappaert, K., Vanderhoeven, J., Van Hum- melen, P., Dutta, B., Clicq, D., Baron, G.V. and Desmet, G. (2003) Enhancement of DNA micro-array analysis using a shear-dri- ven micro-channel flow system. J Chroma- togr A, 1014, 1–9. 73. Hessner, M.J., Wang, X., Khan, S., Meyer, L., Schlicht, M., Tackes, J., Datta, M.W., Jacob, H.J. and Ghosh, S. (2003) Use of a three-color cDNA microarray platform to measure and control support-bound probe for improved data quality and reproducibil- ity. Nucleic Acids Res, 31, e60. Introduction to Microarray Technology 21
- 34. 74. Hessner, M.J., Wang, X., Hulse, K., Meyer, L., Wu, Y., Nye, S., Guo, S.W. and Ghosh, S. (2003) Three color cDNA microarrays: quanti- tativeassessmentthroughtheuseoffluorescein- labeled probes. Nucleic Acids Res, 31, e14. 75. Lindroos, K., Sigurdsson, S., Johansson, K., Ronnblom, L. and Syvanen, A.C. (2002) Multiplex SNP genotyping in pooled DNA samples by a four-colour microarray system. Nucleic Acids Res, 30, e70. 76. Bao, Y.P., Huber, M., Wei, T.F., Marla, S.S., Storhoff, J.J. and Muller, U.R. (2005) SNP identification in unamplified human geno- mic DNA with gold nanoparticle probes. Nucleic Acids Res, 33, e15. 77. Hesse, J., Jacak, J., Kasper, M., Regl, G., Eich- berger, T., Winklmayr, M., Aberger, F., Sonn- leitner, M., Schlapak, R., Howorka, S. et al. (2006) RNA expression profiling at the single molecule level. Genome Res, 16, 1041–1045. 78. Han, A., Dufva, M., Belleville, E. and Chris- tensen, C.(2003) Detection of analytebinding to microarrays using gold nano particles labels and a desktop scanner. Lab Chip, 3, 336–339. 79. Alexandre, I., Hamels, S., Dufour, S., Collet, J., Zammatteo, N., De Longueville, F., Gala, J.L. and Remacle, J. (2001) Colorimetric silver detection of DNA microarrays. Anal Biochem, 295, 1–8. 80. Petersen, J., Stangegaard, M., Birgens, H. and Dufva, M. (2007) Detection of muta- tions in the beta-globin gene by colorimetric staining of DNA microarrays visualized by a flatbed scanner. Anal Biochem, 360, 169–171. 81. Chen, J.J., Wu, R., Yang, P.C., Huang, J.Y., Sher, Y.P., Han, M.H., Kao, W.C., Lee, P.J., Chiu, T.F., Chang, F. et al. (1998) Profiling expression patterns and isolating differen- tially expressed genes by cDNA microarray system with colorimetry detection. Geno- mics, 51, 313–324. 82. Eisen, M.B. and Brown, P.O. (1999) DNA arrays for analysis of gene expression. Meth- ods Enzymol, 303, 179–205. 83. Workman, C., Jensen, L.J., Jarmer, H., Berka, R., Gautier, L., Nielser, H.B., Saxild, H.H., Nielsen, C., Brunak, S. and Knudsen, S. (2002) A new non-linear normalization method for reducing variability in DNA microarray experiments. Genome Biol, 3, research0048. 84. Luebke, K.J., Balog, R.P. and Garner, H.R. (2003) Prioritized selection of oligodeoxyr- ibonucleotide probes for efficient hybridiza- tion to RNA transcripts. Nucleic Acids Res, 31, 750–758. 22 Dufva
- 35. Chapter 2 Probe Design for Expression Arrays Using OligoWiz Rasmus Wernersson Abstract Since all measurements from a DNA microarray is dependant on the probes used, a good choice of probes is of vital importance when designing custom microarrays. This chapter describes how to design expres- sion arrays using the OligoWiz software suite. The desired general features of good probes and the issues which probe design must address are introduced and a conceptual (rather than mathematical) description of how OligoWiz scores the quality of the potential probes is presented. This is followed by a detailed step- by-step guide to designing expression arrays with OligoWiz. The scope of this chapter is exclusively on expression arrays. For an in-depth review of the entire field of probe design (including a comparison of different probe design packages) as well as instructions on how to produce special purpose arrays (e.g., splice detection arrays), please refer to (1). Key words: Probe design, probe selection, expression array, oligonucleotide array, DNA microarray, software, bioinformatics, transcripts. 1. Introduction A good choice of probes is vital to the usefulness of a microarray since the probes determine what signal will be detected (from both intended and non-intended targets). In summary a good probe must fulfill the following criteria: An ideal probe must discriminate well between its intended target and all other potential targets in the target pool. The probe must be able to detect concentration differences under the applied hybridization conditions. OligoWiz website: http://www.cbs.dtu.dk/services/OligoWiz/ Martin Dufva (ed.), DNA Microarrays for Biomedical Research: Methods and Protocols, vol. 529 ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009 DOI 10.1007/978-1-59745-538-1_2 Springerprotocols.com 23
- 36. These two points are the ultimate goal to achieve for all probe design software packages, even if the actual algorithms used can be quite different (1). The following sections describe how this is handled in the OligoWiz software package. 1.1. Introducing OligoWiz Since the computational burden of performing the scoring of all possible probe positions is substantial, OligoWiz has been imple- mented as a client–server solution. The workflow is as follows: The user interfaces with the Graphical User Interface (the ‘‘client’’ – written in Java for platform-independent use,see Fig. 2.1), and selects a dataset and a set of parameters for the probe design project. Next the Fig. 2.1. OligoWiz 2.0 screenshot. This screenshot shows the main functionality of the software – including the graphical representation of the probe-goodness scores and the placement of probes along the selected transcript. The orange bar below the curves represents the currently selected transcript (dnaA). In this example a short-mer (24–26 bp) probe design for Bacillus subtilis is in progress, and up to 15 probes per transcript have been placed. The probes are visualized as lines below the transcript, and details are provided in the lower right-hand corner. Please note that the five scores are color-coded (cannot be seen here) – examples of the color coding is found at the OligoWiz website and in OligoWiz publications (1–3). 24 Wernersson
- 37. data is uploaded to the server (hosted at a multi-CPU super- computer located at the Center for Biological Sequence Ana- lysis at the Technical University of Denmark) where all the computationally heavy algorithmic processing takes place. Once calculation of a particular dataset is completed a datafile with scoring information about each potential probe along all transcripts in the dataset is returned to the user, and all further work on the actual probe selection happens in a completely off- line fashion using the GUI. 1.2. Probe Suitability Scores in OligoWiz OligoWiz uses a scoring-scheme that works as follows: For each position along all transcripts in the input dataset the suitability of placing a probe here is evaluated according to five criteria: Cross- hybridization, Tm, Folding (self-annealing), Position (within the transcript) and Low-complexity. Each individual score has a value between 0.0 (not suited – a bad position for placing a probe) to 1.0 (well suited – no problems detected). The individual scores are then combined with different weights (e.g., Cross-Hybri- dization is more important than Low-Complexity, see Fig. 2.1 for the default values) to form a Total score which is also normalized to be between 0.0 and 1.0. The actual selection of the best position for probe placement is based on the Total score. In the following sections the conceptual workings of the individual scores will be described. The actual formulas for the calculations are found in the two main OligoWiz publications (2, 3). 1.2.1. Cross- Hybridization As mentioned previously a vital property for a probe is to pick up only the intended signal. A way to ensure this is to avoid probes that may hybridize (partially) to other transcripts. It has been shown (4) that a 50-mer will detect a significantly false signal from an unintended target that has more than 75–80% identity at the sequence level. Also, short stretches ( 15 bp) of complete complementarity will give rise to a signal from cross-hybridiza- tion. Similar result for short oligos (23–27 bp) has recently been shown by (1). The perfect way to get around this problem is to calculate the actual hybridization energy between all probes and all targets at the correct individual concentrations. However, since the concentrations of the targets are not known, and since such calculations are very time-consuming we have opted for an approximate solution: screen the entire genome (for prokaryotes and small eukaryotes) or transcriptome (Unigene collection for large eukaryotes, like mammals) using BLAST (5, 6) for regions with substantial similarity to the transcripts in question. By default regions with more that 75% similarity over at least 15 bp is considered to be problematic. Probe Design for Expression Arrays Using OligoWiz 25
- 38. 1.2.2. Tm Another important aspect of probe design is to ensure uniform hybridization conditions throughout the array. Traditionally this has been done by controlling the GC ratio within the probe. OligoWiz addresses this issue by forcing the distribution of Tm (melting temperature) to be as narrow as possible. This is done in two ways: A Tm score1 that evaluates how far the Tm of a potential probe is from the mean Tm of all potential probes. Allowing the length of the probes to vary. Fig. 2.2 shows how the Tm distribution of a set of oligonucleotides becomes increasingly narrow, if the most optimal length (within an interval) can be chosen. Working with short probes it is the experience of the OligoWiz authors that even allowing the length to vary just between 24–26 bp will improve the Tm profile. Finding the optimal length is the very first step performed by OligoWiz: For each position the most optimal length within the user-specified interval is determined, and this length is used for the calculation of all other scores. 1.2.3. Folding To ensure uniform hybridization conditions for all probes on the microarray, the probes should avoid self-annealing (folding). The classical way of investigating this issue is to calculate the free 75 77 79 81 83 85 87 89 91 93 95 97 50bp 52–48bp 54–46bp 56–44bp 58–42bp 60–40bp 0 500 1000 1500 2000 2500 3000 Number of oligos within 1 degree interval Tm Oligo length Tm distribution Fig. 2.2. Tm distribution in optimized length intervals of oligonucleotides. This figure shows how the Tm-distribution of a large set of oligonucleotides (based on all 50 mers within the Yeast genome) can be made increasingly narrow by allowing the length to vary and selecting the most optimal length within each interval. (Based on data from (2) ). 1 Listed as Delta-Tm in the interface. 26 Wernersson
- 39. energy of potential secondary structures using programs such as MFOLD (7). However, using MFOLD is very time consuming2 and for this reason approximate methods that are two orders of magnitude faster was developed for OligoWiz (3). Briefly, this method is based on the idea of aligning the oligo to itself using a dinucleotide alphabet using dynamic programming (8) and a subsitution matrix based on the dinucleotide binding energies. The resulting alignment will represent the lowest folding energy state given the input sequence. This approximate method is in good agreement with MFOLD (see (3), Fig. 2.2) – especially for the sequences with strong secondary structure, which are the most important to avoid when designing probes for DNA microarray. Since all possible probe positions along all target transcripts must be scored, the calculations can be done in a sliding window fash- ion, where most of the dynamic programming matrix from the previous position can be reused, this contributes significantly to the speed-up. Please see (3) for further details on the implementation. 1.2.4. Low Complexity In order to avoid picking up background signal, probes that contain a lot of sub-words that are common in the genome/ transcriptome should be avoided. This can be illustrated with the following example (human DNA): Oligo with low-complexity: AAAAAAAGGAGTTTTTTTTCAAAAAACTTTTTAAAAAAGCTTTAGGTTTTTA Oligo without low-complexity: CGTGACTGACAGCTGACTGCTAGCCATGCAACGTCATAGTACGATGACT In OligoWiz, this problem is addressed by counting the occurrence of all 8 bp words in the genome/transcriptome and scoring the degree to which a probe consist of frequent sub- words. 1.2.5. Position The optimal position within the transcript for placing a probe depends on the labeling and/or amplification method used. When using standard poly-T priming (targeting the poly-A tail of eukaryotic transcripts) the labeling starts from the 30 end of the transcript. Since there is a certain probability that the reverse transcriptase will not complete the synthesis of cDNA in full length, most signals are detected using probes targeting the 30 end of the transcript. In OligoWiz the following position prefer- ence models are built in. 2 2 seconds for a 30-mer and 16 minutes for all 30-mers in a 500 bp transcript at OligoWiz reference platform at the time. Probe Design for Expression Arrays Using OligoWiz 27
- 40. Poly-T priming: Push probes towards the 30 prime end (Probabilistic model of the labeling from the 30 prime end). Random priming: Avoid probes at the extreme 30 prime end (Probabilistic model of the labeling using random hexamers). Linear 50 preference: 1.0 at the 50 end and decreases linearly to 0.0 over 2000 bp. Linear 30 preference: As the 50 preference, but counting from the 30 end instead. Linear mid preference: 1.0 at the midpoint decreasing to 0.0 over 1000 bp to each side. Observe that it is possible to completely ignore the position score, by setting its weight to 0.0. This is especially useful in situations like placing splice-junction probes, where the position is constrained by the gene structure. 1.3. Rule Based Placement of Probes As mentioned previously, the OligoWiz server returns a datafile to the client (the graphical interface) which contains scoring of all possible probes. At this point no decisions about the actual placement (how many per transcript, spacing etc.) of the probes have been made. All the computations on the place- ment of the probes is performed solely on the user’s own computer in a completely off-line manner. This means that once the data file has been created it contains everything needed for further work, and can be stored on the user’s own computer/network or be shared with collaborators using email, for instance. The actual placement of the probes is done using a rule based method (see Fig. 2.3 for an overview of the options). The placement algorithm is as follows (repeated for each transcript): 1. Apply filters: If any filters have been defined (e.g., requiring the total-score to be above a certain value), start by masking out the regions disallowed by the filters. For the advanced optional use of filter please see (1). 2. Place probe: Select the currently available position with the highest Total score for probe placement. 3. Mask out surrounding positions: Positions within the desired minimum distance are masked out. 4. Repeat/terminate: Terminate if the maximum total number of probes has been reached or if no more positions are avail- able. Otherwise, go to step 2. Since the computationally heavy calculations (scoring of all probe position) have already been performed on the server, the placement algorithm is fast. This makes it possible to 28 Wernersson
- 41. experiment with the probe placement parameters, evaluate the result, and refine the parameter in a real-time iterative fashion. 1.4. Exporting the Probe Sequences The final step in the probe design process will be to actually order the array (e.g., NimbleExpress) or the oligonucelotides to be spotted. In order to make this step easy, OligoWiz support exporting the probe sequence to both FASTA and TAB format, and has the option of reverse-complimenting the probes (if needed) and automatically creating PM/MM probe pairs, if that is desired. Furthermore, it should be noted that a Material and Methods section describing the parameters used in the probe design is auto-generated and added to the file, Fig. 2.3. Probe selection dialog. The spacing criteria are specified in the topmost box. The use of filters and sequence feature annotation (e.g., intron/exon structure) are not described here. For further details please refer to the OligoWiz website and (1). Probe Design for Expression Arrays Using OligoWiz 29
- 42. documenting the probe design process (see Fig. 2.4 and step 10 in the step-by-step guide). 2. Materials An internet connected computer with Java 1.4 (or newer) installed. The OligoWiz client is tested on Windows, Mac OS X, Irix, Solaris and Linux – it is written with cross-platform use in mind and should work on virtually any operating system for which a Java Runtime Environment (JRE) exists. The optional Fig. 2.4. OligoWiz probe sequence export options. 30 Wernersson
- 43. use of a local installation of the OligoWiz server software is not covered here, please see (1) and the OligoWiz website for further details. 3. Methods This section summarizes the steps the user has to go through to select probes for an expression array. 1. Prepare target sequences in FASTA format. (For instruc- tions on how to use TAB files please see (1, 9) – or the descriptions on the OligoWiz website). The very first step is to identify the sequences that the array should detect. This could for example be an entire prokaryotic genome or a set of transcripts from the human genome/transcriptome. For prokaryotic sequences a file prepared from the CDS (protein coding genes) regions of the full genomic sequence is recommended. In many cases a FASTA file with only the transcripts/CDSs can be downloaded from the same data- source as the full genomic builds. For higher eukaryotes (e.g., Human or Mouse), sequences from the UNIGENE collec- tions are recommended. Observe that it is important to also include control targets/genes – since most normalization algorithms used in the downstream processing assumes that only a minor (10%) proportion of the transcript vary from array to array (10). See Note 1 for further details about the input data. 2. Launch the OligoWiz client. 2.1. Download the most recent version of the OligoWiz client from the OligoWiz website: www.cbs.dtu.dk/services/ OligoWiz/. 2.2. Download Java version 1.4 (or newer) if it is not already installed on the local computer. Instruction on how to do this on various platforms (Windows/Linux/Mac) is detailed on the webpage. 2.3. Launch the OligoWiz client by double-clicking on the JAR file (Windows and Mac) or from the command-line (Linux and UNIX). See Note 2 for issues relating to the memory usage of the program. 3. Select input file. Click the ‘‘...’’ button next to the Input FASTA or TAB file field (see Fig. 2.5), and select the FASTA file prepared in Step 1. The OligoWiz client will suggest a unique filename for the result file (not generated yet) – accept this, or customize the filename/placement if desired. Probe Design for Expression Arrays Using OligoWiz 31
- 44. 4. Select species database. Select the species database that will be used for calculating the Cross-hybridization and Low- Complexity scores. A full description of all the databases3 is available on the OligoWiz website. (If the species-tree is empty, please refer to Note 3 describing how to trouble- shoot network issues). 5. Customize score parameters. Select the best fitting prede- fined parameter set in the Score parameters/info box and press Load (see Fig. 2.5). The predefined parameter sets can be customized further, as described below: 5.1. Oligo Length: Determines if OligoWiz should aim at a fixed length or allow the length to vary within an interval in order to optimize Tm (recommended). 5.2. Tm 5.2.1. Select if OligoWiz should determine the optimal Tm (recommended) – alternatively a specific Tm to aim for can be specified. Fig. 2.5. OligoWiz query launch page. 32 Wernersson
- 45. 5.2.2. Select if OligoWiz should use a DNA:DNA or RNA:DNA model for calculating the Tm. Select DNA:DNA if DNA is to be hybridized to the array and RNA:DNA if RNA is used (this is typically the situation). 5.3. Cross-Hybridization 5.3.1. Set the cut-off values of when a BLAST hit is to be considered: % minimum similarity and mini- mum length. Hits below this threshold will be completely ignored. It is recommended to use the default values. 5.3.2. Set the cut-off when a BLAST hit is considered a ‘‘self-hit’’ (the target sequence it self). For prokaryotic arrays the default values are recom- mended – if the input data is transcripts for a complex eukaryotic organism with a large degree of alternative splicing, the issue of detecting self- hits is more complicated. In this case it is recom- mended to lower the self-hit criteria. A pragmatic solution is to lower the self-hit length criteria to 40% (0.4) – see (1) for a detailed discussion. 5.4. Select position model. For labeling protocols using poly-T (usually the case for running eukaryotic arrays) select the Poly-T option. For labeling protocols using random hexamers (usually the case for prokaryotic arrays) select the Random priming option. 6. Submit the query 6.1. Optional step: Enter your email address in the Email address field – this will make the server send you an email once the processing is completed with a link to direct download of the result data file. This is especially useful for long running queries. 6.2. Press the ‘‘Submit’’ button 7. Wait for the server to finish processing the query. The status of the processing can be seen in the Query List table. Once the processing has completed, the data file (file type: .owz.gz) will automatically be downloaded and stored on the local computer. 8. Load the data file. Double-click on the downloaded query in the ‘‘Query List’’ table to load in the data. This will load in the data and launch the main interface for placing probes (see Fig. 2.1). Notice: If the data file has been downloaded manually by following the link in the server-generated email, the data can be loaded by using the File - Open menu option. Probe Design for Expression Arrays Using OligoWiz 33
- 46. 9. Place probes 9.1. Adjust score weights (if needed). It is recommended to keep the default settings. However, notice that it’s possible to disable a score by setting its weight to 0.0. 9.2. Bring up the Oligo Placement window. Press the ‘‘Place Oligos...’’ button to launch the probe selection dialog (see Fig. 2.3). 9.3. Select probe placement criteria. For short probes (25 bp) 8 probes or more per target sequence is recom- mended, for long probes (50–70 bp) 2–4 (or more) is recommended (1). 9.4. Apply selection criteria. Press the Apply to all button to search for probes fulfilling the criteria in the entire data set. (The Apply button can be used to test the criteria on a single sequence). 9.5. Inspect the placement of the probes. Keep the probe placement window open, and inspect the placement of the probes in the main window. Notice that both the Entries and Oligos lists can be sorted by clicking on the header elements. This makes it easy to identify target sequences for which no or few probes have been selected. 9.6. Repeat step b-e if needed. 10. Export probe sequences. Press the‘‘Export oligos...’’buttonto bring up the Probe Export window (see Fig. 2.4). The sequences can be exported in FASTA and TAB format. Optionally the probe sequences can be exported as anti-sense probes and/or pairs or PM/MM (perfect match/Mis-match) probes can be generated. In most cases the probes should be saved as ‘‘sense’’ probes in FASTA format – however, it is important to make sure that the strandness is correct for the protocol to be used in the lab. 11. Optional: Export negative set. If a sub-set of the target sequence proves to be difficult to design probes for, this sub-set can be extracted from the full set of target sequences, by pressing the Export negative set button. This makes it possible to isolate the troublesome cases, and re-run the entire probe-design process for these sequences only with more relaxed settings (or alternatively deciding NOT to tar- get these sequences in the array design). 4. Notes 1. Problems related to input data: The most common source of problems with running OligoWiz is problems with the input data: 34 Wernersson
- 47. 1.1. Please make sure that the data is in a supported file format (TAB or FASTA). Notice that the file must be a text-only file (an otherwise correctly formatted FASTA file within a MS-Word document will NOT work). 1.2. Please make sure that the file contains the sequences of the transcripts/genes which should be targeted. Submitting a file with a single large DNA sequence representing an entire prokaryotic genome will not work. OligoWiz is designed to work in a gene/transcript oriented way (for comments on how to design a chro- mosomal tiling array please see (1)). 1.3. Please make sure that the input sequences are of a suffi- cient length. Entries that are shorter than the minimum probe length will be discarded. 2. Memory problems: For very large datasets, the default amount of memory available to Java may become a problem. As a rule of thumb more memory may be needed if a FASTA file with more than 10,000 sequences (average prokaryotic CDS length) is submitted. The OligoWiz webpage contains detailed instruction of how to start the OligoWiz client with more memory on various platforms. 3. Network problems: If the OligoWiz client fails to connect to the OligoWiz server (the species database list remains empty, and the connection status remains ‘‘not con- nected’’) it is most likely to be due to problems with the network setup. The OligoWiz client communicates with the server using HTTP (like a web browser), and it needs a direct connection rather than going through a HTTP proxy. If the local network setup uses a HTTP proxy (inspect the browser proxy settings – or ask the local system administrator), this is likely to be the cause of the problem. The OligoWiz website contains a descrip- tion of a work-around of this issue. References 1. Wernersson, R., Juncker, A.S. and Nielsen, H.B. (2007) Probe Selection for DNA Microarrays using OligoWiz. Nature Proto- cols, 2, 2677–2691. 2. Nielsen, H.B., Wernersson, R. and Knudsen, S. (2003) Design of oligonucleotides for microarrays and perspectives for design of multi-transcriptome arrays. Nucleic Acids Res, 31, 3491–3496. 3. Wernersson, R. and Nielsen, H.B. (2005) OligoWiz 2.0-integrating sequence feature annotation into the design of microarray probes. Nucleic Acids Res, 33, W611–W615. 4. Kane, M.D., Jatkoe, T.A., Stumpf, C.R., Lu, J., Thomas, J.D. and Madore, S.J. (2000) Assessment of the sensitivity and specificity of oligonucleotide (50 mer) microarrays. Nucleic Acids Res, 28, 4552–4557. 5. Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) Basic local alignment search tool. J Mol Biol, 215, 403–410. Probe Design for Expression Arrays Using OligoWiz 35
- 48. 6. Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W. and Lip- man, D.J. (1997) Gapped BLAST and PSI- BLAST: a new generation of protein database search programs. Nucleic Acids Res, 25, 3389–3402. 7. Zuker, M. (1994) Prediction of RNA sec- ondary structure by energy minimization. Methods Mol Biol, 25, 267–294. 8. Needleman, S.B. and Wunsch, C.D. (1970) A general method applicable to the search for similarities in the amino acid sequence of two proteins. J Mol Biol, 48, 443–453. 9. Wernersson, R. (2005) FeatureExtract- extraction of sequence annotation made easy. Nucleic Acids Res, 33, W567–W569. 10. Workman, C., Jensen, L.J., Jarmer, H., Berka, R., Gautier, L., Nielsen, H.B., Saxild, H.-H., Nielsen, C., Brunak, S. and Knudsen, S. (2002) A new non-linear normalization method for reducing variability in DNA microarray experiments. Genome Biol, 3, research0048. 36 Wernersson
- 49. Chapter 3 Comparative Genomic Hybridization: Microarray Design and Data Interpretation Richard Redon and Nigel P. Carter Abstract Microarray-based Comparative Genomic Hybridization (array-CGH) has been applied for a decade to screen for submicroscopic DNA gains and losses in tumor and constitutional DNA samples. This method has become increasingly flexible with the integration of new biological resources generated by genome sequencing projects. In this chapter, we describe alternative strategies for whole genome screening and high resolution breakpoint mapping of copy number changes by array-CGH, as well as tools available for accurate analysis of array-CGH experiments. Although most methods listed here have been designed for microarrays comprising large-insert clones, they can be adapted easily to other types of microarray plat- forms, such as those constructed from printed or synthesized oligonucleotides. Key words: Probe design, clone selection, normalization, outlier detection, CNV calling, Comparative Genomic Hybridization, array-CGH. 1. Introduction Comparative Genomic Hybridization (CGH) was developed in the early 1990s to screen for chromosomal deletions and duplica- tions along whole genomes (1, 2). Originally, CGH consisted of co-hybridizing one test and one reference labeled probe DNA onto metaphase chromosomes spread on glass slides in the presence of Cot-1 DNA to suppress high repeat sequences (see Chapter 17). During the 1990s, CGH on chromosomes was widely used by research laboratories, in particular to screen for chromosome numerical aberrations associated with the progres- sion of solid tumors (3): chromosome analysis by G-banding was Martin Dufva (ed.), DNA Microarrays for Biomedical Research: Methods and Protocols, vol. 529 ª Humana Press, a part of Springer ScienceþBusiness Media, LLC 2009 DOI 10.1007/978-1-59745-538-1_3 Springerprotocols.com 37
- 50. technically challenging with tumor cells, due to the frequency of highly rearranged karyotypes and difficulties in culturing cells in vitro to obtain good quality metaphase chromosomes. However, although CGH became widely used in cancer research, it did not prove to be particularly valuable as a standard method in diagnostic laboratories for the analysis of genomic imbalance in patients with developmental disorders. This was firstly due to the poor spatial resolution of metaphase CGH, which is limited to 5–15 Mb by the image acquisition of probe signals on metaphase spreads using fluorescence microscopy. Secondly, metaphase CGH is technically challenging, requiring expertise for preparation of suitable metaphase chromosomes as well as image acquisition and analysis. From the mid 1990s, the International Human Genome Sequencing Project released new information on the human gen- ome sequence, which was derived from the construction and characterization of libraries comprising large-insert clones such as bacterial artificial chromosomes (BACs) (4). These resources allowed the CGH method to be modified such that metaphase chromosomes could be replaced by arrayed DNA fragments representing precise chromosome coordinates. This strategy was initially called matrix-CGH (5) and then array-CGH (6), and it is this name that is now in common usage. The development of array-CGH improved significantly the potential of CGH for the analysis of small chromosomal imbalances. Initial arrays provided a more than tenfold increase in resolution such that micro rear- rangements that were invisible previously on chromosome pre- parations became detectable. Also, for the first time, deletion and duplication breakpoints could be localized directly on the human genome sequence assembly. The large insert clones used for the first array-CGH applica- tions – in particular BACs and fosmids – have since become widely available. This has facilitated the construction of microarrays cov- ering the whole genome at increasingly higher resolution. How- ever, the relatively large size of these clones (170 kb for BACs, 40 kb for fosmids) limits the ultimate resolution of these types of arrays. In the past couple of years, small-insert clones, PCR products, and oligonucleotides have been developed for use in array-CGH (7, 8) allowing a greater degree of flexibility and higher resolution (down to just a few base pairs) in the design of microarray experiments, which can be tailored to the specific biological question. This chapter describes many critical factors that should be considered when designing new array-CGH experiments and discusses different possible strategies for data analysis. It focuses on microarrays comprising cloned DNA printed on slides, though some strategies and tools described here can also apply for the design of microarrays composed of printed or synthesized oligonucleotides. 38 Redon and Carter
- 51. 2. Array-CGH Design 2.1. Clone Selection The first step in array-CGH is the design or choice of the micro- array to be used for interrogating test genomes. There are two common strategies: (i) the design or the selection of one micro- array covering the whole genome in order to screen for every deletion or duplication in a given test genome compared to a reference DNA; (ii) the construction and use of one microarray targeted to one part of the genome only, such as one chromosome or one region. The design of a whole genome microarray is dependent on the resources available to construct the array. Construction of arrays from large insert clones requires physical spotting of the clone DNA onto microscope slides, which typically limits the number of elements on the array to less than 50,000. For this reason, many laboratories used BAC clones for whole genome coverage, because with an average length of 170 kb coverage of the whole genome with overlapping clones requires approximately 30,000 BACs while it would require more than 120,000 fosmids (40 kb in length). Covering the whole genome at tiling path resolution is an important investment in time and resources, which may not be suitable for many laboratories. For this reason, most BAC micro- arrays used for whole genome screening comprise only approxi- mately 3,000 clones. They cover the whole genome with clones regularly interspaced, each single clone positioned at an interval of approximately 1 Mb apart. Although this strategy is not efficient for the detection of copy number changes below 1–2 Mb in size, it has proved to be valuable for the screening of most large-scale deletions or duplications, such as those responsible for severe congenital anomalies. Several sets of clones designed specifically for the construction of CGH microarrays are publicly available. The Wellcome Trust Sanger Institute has developed two sets of large-insert clones for the construction of microarrays covering the whole genome at 1-Mb and tiling path resolutions (1Mb and 30k TPA sets, respectively). The coverage of the human genome by these two sets of clones can be visualized on the Ensembl browser (www.ensembl.org, see Fig. 3.1A) and clones are available through GeneService (www. geneservice.co.uk). Another selection of 32,000 over- lapping BAC clones covering the whole genome can be obt- ained from the BACPAC Resources Center at CHORI (bacpac. chori.org). To design a microarray targeted to specific loci, there is a larger choice of clones which could be used, depending on the size of the genomic segments to cover and on the resolution which is required. While BAC clones are usually selected for the construction of whole-genome microarray, fosmid clones Comparative Genomic Hybridization 39
- 52. represent a good alternative for custom arrays. Overlapping fos- mids provide better resolution than overlapping BACs (down to 10 kb in case of high redundancy in coverage versus approximately 50 kb) but can be prepared for spotting using the same protocols (see Chapter 16). The fosmid library WIBR-2 is particularly useful as it has been extensively characterized by end-sequencing: most clones from this library are precisely mapped on the human gen- ome assembly and all read-pair positions can be visualized on the UCSC genome browser (genome.ucsc.edu, see Fig. 3.1B). Read- pair coordinates can be downloaded from the UCSC browser for further selection of the clones required to cover the regions of interest. All fosmids can be purchased at the BACPAC Resources Center (bacpac.chori.org). For example, after selecting fosmids for the construction of a small custom microarray, we applied array-CGH for high-resolution breakpoint mapping of two deletions at 9q22.3, responsible for a syndrome involving mental retardation and overgrowth in two unrelated children (9). The result obtained for one child is Fig. 3.1. Selection of large-insert clones for array-CGH using Genome Browsers (A) The Ensembl browser (www.ensembl. org) enables the user to visualize many physical or biological annotations in the context of the genome sequence. The box displays the respective positions of genes (Ensembl annotation, top panel), clones from the Sanger 1Mb set (middle panel) and clones from the 30k TPA set (bottom panel) between coordinates 95–100 Mb on human chromosome 9. Lists of clones from these two sets can be downloaded as delimited tables from the same website (select option ‘‘Graphical overview’’). (B) Part of the same interval (99–100 Mb), displayed on the UCSC Genome Browser (genome.ucsc.edu), one alternative to Ensembl. The bottom panel shows positions of clones from the 30k TPA set. The top panel displays the positions of many fosmids mapped by pair-end sequencing. Some of the fosmid clones can be selected by their chromosomal locations for high-resolution coverage of the locus by array-CGH. 40 Redon and Carter
- 53. shown in Fig. 3.2A. Further increase in array-CGH resolution can be achieved by selecting small-insert clones (1.5–4 kb, see Fig. 3.2B) or PCR products (less than 1 kb), which can be used to cover all exons of any gene of interest (7). Today, synthetic oligonucleotides have largely replaced these approaches to custom Fig. 3.2. High resolution breakpoint mapping by array-CGH (A) Array CGH profiles at the proximal (left) and distal (right) breakpoints of a 9q22.3 deletion detected in a patient with overgrowth syndrome (9). The deletion was first detected with a microarray covering the whole genome at 1 Mb resolution (positions of 1 Mb clones are represented as large grey bars). One custom microarray comprising fosmids (represented as short black bars) was then constructed to cover the two breakpoint regions at tiling path resolution. CGH with the custom array refined the deletion breakpoints to intervals of less than 50 kb. Note that the 1 Mb array profile was normalized by a block median method, while the custom array was normalized by the median of log2ratios from 26 fosmids located on chromosome 18 and used as controls (5). (B) Detailed views of the same deletion breakpoint intervals. Using a small custom microarray comprising small-insert clones (1.5 to 4 kb in length, represented as small grey bars), it was possible to map each deletion breakpoint at a resolution of less than 5 kb. Long-range PCR amplification and sequencing confirmed that array-CGH applied with increasing resolution enables accurate mapping of deletion breakpoints. The actual breakpoints are shown below the profiles on the UCSC browser: the proximal breakpoint disrupts the first intron of the PHF2 gene while the distal breakpoint is distal to the NR4A3 gene. Comparative Genomic Hybridization 41
- 54. array construction. Several companies – such as Agilent Technol- ogies, Inc. and NimbleGen Systems, Inc. – are now commercializ- ing microarray platforms with custom oligonucleotide synthesis, which provides virtually unrestricted flexibility in the design of CGH. 2.2. Controls The microarray design should always include a selection of control target sequences, which will be used to estimate the performance of the microarray as well as the quality of array- CGH hybridizations. Some negative controls should be included to estimate the intensity of fluorescence resulting from the non-specific hybridi- zation of genomic probes on the target DNA. For printed arrays, negative control spot positions commonly contain bacterial geno- mic DNA or DNA sequences from other species, such as Droso- phila. After image acquisition and spot intensity quantification, the intensity of fluorescence on these negative controls should always be monitored and be extremely low when compared to the test intensities along the microarray. It is also valuable if possible within the array design to include controls for the estimation of the dosage response on the array. For example, adding clones representing sequences on chromo- some X can be used to estimate the ratio deviation due to the presence of one copy in a male test DNA compared to 2 copies in a reference female DNA. This strategy has been widely used to validate the performance of new microarray platforms (6, 7). In addition, it may be useful to include some normaliza- tion probes particularly for microarrays covering only small regions. Selecting a number of clones that are located in one or several regions of the genome unlikely to be variable in copy number in test and reference DNA samples can be critical for normalization steps (see Fig. 3.2). The control clones can either be located on a chromosome which is known to contain no gross anomaly or can cover genes which are known to be present in normal copy number in the test and the reference DNA. When working on copy number variations (CNV) in humans, one common strategy consists in selecting only clones located at chromosomal loci not reported to show variation in the literature (data available in the Database of Genomic Var- iants, projects.tcag.ca/variation). At last, using one or a small group of clones that will be printed in replicate distributed regularly on the surface of the microarray can help in detecting problems of signal heterogeneity after hybridization and imaging. Furthermore, a control DNA sequence spotted in replicate along the array can be used to estimate and correct the spatial heterogeneity of log2ratio values (see Section 3.1.3). 42 Redon and Carter
- 55. Other documents randomly have different content
- 56. TONY DREW CLOSER TO LISTEN He would have taught him to be a loyal Italian. For Anna's father was a real patriot. Robert Browning, the poet, has said, Open my heart and you will see inside of it—Italy. If Anna's father had been a poet, he might have said something like this. Dinner is ready, announced Anna's mother. Tony watched as the family left the room. He knew that they had gone into the dining room. He waited patiently beneath the window until they returned.
- 57. When they came back, Anna's father eased himself into an armchair. Come, little Anna, he said. I am going to read to you. Anna crawled on to his lap with Tina clasped lovingly in her arms. Tina had a puffed, happy look, as if she, too, had dined well! Tony smiled to himself. He was going to hear Anna's father read stories. No one had ever read to Tony. He loved reading. The night was warm. The moon shone. The window was open. Tony listened. Would you like to listen, too? Very well. Wouldn't Anna's father be surprised if he knew about his big audience? Under the window is a poor Italian boy—Tony. Out in the great United States are other boys and girls—you who are reading this tale! So be very quiet and don't make a noise for fear of disturbing Anna's father while he reads. Let us crouch under the window with Tony! CHAPTER IV
- 58. ROME Tonight, began Anna's father, we are going to read about one of our Italian cities. Many fine stories have come out of it. Rome is called 'The Eternal City' because there is a saying that it will live forever. It is built upon seven hills. A long time ago there lived a great artist named Michelangelo. He built the dome of St. Peter's Cathedral in Rome. This is the largest church in the world. Thirty services may be conducted in it at the same time. The bones of St. Peter are believed to have been buried beneath the Cathedral.
- 59. ST. PETER'S: ROME But the oldest church of all is the Pantheon, which means 'all the Gods,' It was built when people worshipped more than one God. It has no
- 60. windows but only a hole in the top called an 'eye.' Today it is the burial ground of renowned writers and artists. THE PANTHEON: ROME Near Rome are the famous catacombs. It was here that the early Christians buried their dead.
- 61. THE VATICAN: ROME The catacombs are long, narrow passages with graves built into the walls, one above the other. When the Christians were not allowed to worship in their own way, they often fled to these underground cemeteries to pray. There is a curious park in Rome, went on the father. One which you, little Anna, would like. Anna looked up. Why, Papa? she asked.
- 62. Because it is filled with cats, answered her father. Tabbies and Tommies, black and white, grey and yellow. They wander about and sprawl in the shade of fine old trees. They have plenty to eat and nothing to fear. It is a kitty paradise! I want to go to that park some day, said Anna. There is a magic fountain in Rome, read her father. It is said that he who drinks from the Fontana Trevi will some day be drawn back to The Eternal City. The Appian Way is sometimes called The Queen of Roads. It was a great highway built by the ancient Romans. Parts of it are still in use. These ancient Romans were very clean. They dotted their city with many fine public baths. We are able to see by the ruins how very handsome they were.
- 63. THE COLOSSEUM: ROME Outdoor theatres, called 'circuses,' were also numerous. The oldest of these is the Circus Maximus, where races were held.
- 64. INSIDE THE COLOSSEUM: ROME The Colosseum is a huge outdoor arena where slaves and criminals were thrown to hungry lions. People sat about and enjoyed the show.
- 65. Of course the poor men were killed. But the audience watched this terrible sport as naturally as we, today, watch a tennis game. They pitied the victims no more than we pity the tennis balls! Anna squirmed unhappily. Now read something nice, she said. The story of Romulus and Remus, because I like the good wolf. Her father smiled and turned a page. Always stories about animals for little Anna! Here we are, he said. The old myth goes that Romulus and Remus were twin babies, cast upon the River Tiber by a jealous king. Their basket floated ashore and was found by a mother wolf. Taking pity on the babies, she brought them to her cave and cared for them. But at last the good wolf was killed by hunters and Romulus and Remus, now grown boys, ran away.
- 66. TREVI FOUNTAIN: ROME A herdsman found them and gave them a home. They were very wild and strong and they were wonderful hunters. One day they learned the story of their lives. They discovered that they were really meant to be kings. So they determined to punish their enemy and take their rightful place in the world. Remus was killed in battle, but Romulus became the first king of Rome.
- 67. The legend tells that, at this time, there were very few women in Rome. Romulus wished his people to marry women of the neighboring cities. But the neighbors refused to marry the Romans. So Romulus invited a people called The Sabines to a great feast. During the entertainment the Romans seized the young Sabine women and carried them off. Later, however, this savage act was forgotten and the two nations became one. In 218 B.C. Rome suffered a defeat. Hannibal, a great general of ancient Carthage, crossed the tall Alps and attacked the Romans. His army consisted of 90,000 foot soldiers, 12,000 horsemen, and 37 elephants. This march over the Alps is considered one of the most wonderful military feats of ancient days. A PARADE PASSING THE COLOSSEUM: ROME
- 68. Nero was one of the most wicked emperors who ever ruled Rome. In the year 64 a terrible fire broke out. For six days flames swept the city. Yet Nero made no attempt to stop the fire nor to help the suffering people. Some say that the cruel Emperor played upon his fiddle while Rome burned. After the World War there came to Rome a new kind of King. He was really not a king at all but.... Il Duce! (The Commander!) interrupted Anna. Yes, my dear, agreed her father. His name was Benito Mussolini, the great chief of Italy. Mussolini was a poor boy, the son of a blacksmith. Like wicked Nero, he sometimes played upon his fiddle. But while he played, Rome did not burn. It grew. He founded a new system of government called Fascism. A wise man once was asked, 'What is the best quality for a child to have?' He replied, 'Obedience,' 'And the second best?' 'Obedience,' 'And the third?' 'Obedience!' This is what the Fascist teachers believe. Their moral is: 'Be strong to be pure. Be pure to be strong,' Il Duce has taught our people this wonderful lesson. At one time there were many lazy ones in Italy. Now we work and clean and teach. It is better that way. Italy is a beautiful land. It is worth working for. Tony, under the window, felt a great pride in his heart. He began to see ahead into the future when he would be an Italian soldier. He would fight for beautiful Italy! He waggled his head back and forth against the side of the house. He muttered to himself, Viva Italia! (Hurrah for Italy!) Viva! Viva.... Ouch! he cried suddenly. He had bumped his head!
- 69. CHAPTER V
- 70. TONY AND ANNA Did I hear a noise outside? asked Anna's father. Anna hugged Tina. It must have been a little mama animal putting its babies to bed, she said. Her father sighed. Some day Anna would be a little mama herself. That was what Mussolini wanted all of Italy's women to be. But Anna's father would so have liked a son. One who would be more interested in the Balilla than in little mama animals. Yet he loved his daughter very dearly. He now kissed her dark curls as he said, It is time for bed, mia cara (my dear). Tomorrow night more stories. Anna sat up in his arms. Tina awoke and blinked. Before I go to bed, I must put Niki to bed, too, said Anna. Her father answered, Then we must make a house for her. Tony saw him open a chest of drawers and take out some curious things. Now, he said to his daughter, Come into the back garden, and we shall see what kind of house-builder I am! Tony watched them leave the room and saw a light switch on in the hallway. Then the back door opened. Father, daughter, and dog went into the garden. They found an old crate with the top missing. They covered it with what appeared to be a fancy tablecover. They tied the little dog securely to the side. There! said Anna's father. It looks like a tent on the desert. Niki will feel like an Arabian Princess!
- 71. AH. TINA MIA, I HAVE FOUND YOU AGAIN. Anna stooped down and caressed her pet. Felicissima notte, Niki, said Anna. This meant Happiest night, Niki, and it is what the Italians say for Good-night. When Anna and her father had left, Tony ran over to the kennel-tent. Tina nearly wagged herself to pieces with joy. Tony sank down beside her. He buried his head in her soft hair.
- 72. Ah, Tina mia (my Tina)! he said. I thought they had taken you from me forever! But I have found you again. He started to untie the dog. He would run away with her. Far away! Never back to Guido! Guido was a thief. A man who stole little dogs! Then, suddenly, Tony remembered that he, too, was about to steal a little dog! He, too, would be a thief if he did that. Tina did not belong to him. She belonged to little Anna. But how could he bear to leave Tina? A tear ran down his cheek. Tina licked it sadly. She seemed to know what he was thinking about. How he longed to snuggle up close to the little dog and go fast asleep. Just as he had done every night since he went to live with Guido.
- 73. ANNA Why did Anna have to love Tina, too? He would stay. Just tonight. He would crawl into Tina's tent with her. In the morning he could decide what to do. He was so sleepy now. He yawned, brushed his tears away, and wriggled into the tent. He curled up in there, with Tina in his arms. But just as sleep came creeping over him, a sound was heard in the garden. Tony gave a start and opened his eyes. Tina gave a low growl.
- 74. Tony looked out and saw a white figure approaching the tent. It was Anna. She was coming back to see her new-found Niki once more. She would find Tony there. She would tell her father! What should he do? His heart began to thump. He lay quite still. Niki! whispered Anna, softly. Silence. Niki! repeated Anna. I have come to kiss you good-night. Here, Niki! She bent down in front of the tent and looked in. It was dark inside. Tony lay flat on the floor and kept very quiet. Anna put her hand inside the tent and felt for her pet. Tina tried to hide from the hand, but it found her and lifted her out tenderly. Anna caressed the dog and spoke gently to her. Now, Niki, she said. You shall go back to bed and mama will cover you up. She had brought with her a doll's blanket. She put Tina back into the tent and tried to make her lie down flat. She could do this so easily with her dolls. But, somehow, Tina was different. Tina did not want to lie down flat! The real reason for this was because Anna was spreading Tina on Tony's face! The poor dog struggled and kicked. The poor boy tried his best to lie still and make no noise. But would you enjoy a dog plastered upon your face? So Tony wriggled. He snorted. He sneezed. Anna saw. She heard. She started and gave a little cry. Tony's head came out of the tent like a turtle's head coming out of its shell.
- 75. HUSH, SAID TONY Hush! said Tony. Anna drew back. Who are you? she gasped. I'm Tony, he replied. Please let me stay here with Tina tonight. Tomorrow I'll go away. Then Anna recognized him. Oh, she exclaimed. You are that naughty Marionette boy who told a lie! I am going to call my father!
- 76. She turned toward the house but Tony quickly caught her arm. No, no! he pleaded. I mean no harm. I love the little dog. Let me stay. Only one night. Do not tell your father—please! In the moonlight Anna could see that tears filled his eyes. She began to feel sorry for him. Are you a very poor little boy? she asked, innocently. Oh, yes, very, very poor! he moaned. I have no home. No mother. No father. Everyone is cruel to me. The little dog was my only friend, and now you have taken her from me. AMALFI Tony was born with the Italian gift for beautiful acting. He now acted his best for Anna. While some of his pitiful tale was true, some was sprinkled with the fairy dust of fancy. Every morning Guido beats me, he made up. He uses a big stick. And when he stops beating me, he makes me sing to him. Then, all day long he feeds me bird-seed mixed with soap and nothing else!
- 77. Anna's gentle eyes grew wider and wider, her tender heart softer and softer. Tony warmed to his work. His success encouraged him. He began to gesture with his arms. He began to invent wild tales. Often I fall upon the streets because I am so hungry, he continued. When it rains, Guido makes me lie outside the whole night through. One morning, when I awoke, I found myself in a pool of water. I had to swim all the way home!
- 78. TONY BENT LOW AND KISSED HER HAND The little girl's lip began to tremble. This gave Tony added courage. He drew a deep breath. His style improved. And once I was thrown over a cliff. Lions came prowling....
- 79. He stopped, for little Anna had begun to cry. Oh, you poor boy! she sobbed. I am so sorry for you! I shall tell my father and mother. They will take care of you. No, you must not do that, said Tony, quickly. If your father knows I am here, he will discipline me! But my father is good, said Anna. That is why he will discipline me, replied Tony. Because I am bad. Now, to a very little girl like Anna, that seemed sensible enough. She believed what Tony told her. She even believed that her father might not be kind to the beggar boy. Often her father was very severe. So she promised that she would not tell. You may stay here every night, poor little boy, she said. I will bring food and leave it for you in a dish. I will put a soft cushion inside the tent. I will never tell my father that you are here. Ah, grazie signorina (thank you, Miss), said Tony, charmingly. He smiled and showed his white teeth. How kind you are! And will you also put some candy on the dish? Yes, I will, poor little boy, she answered. What kind do you like? Tony thought a moment. Then he replied, Torrone. (This is the finest and most expensive Italian candy.) Anna promised to leave some torrone. Tony bent low and kissed her hand as he had seen the marionettes do in romantic plays. Felicissima notte, bella signorina! (Good-night, beautiful Miss!) he murmured. Again his play acting and falsehoods had brought him success! He did not even know that he had done anything wrong. Poor neglected little Tony! CHAPTER VI
- 80. CITIES, ANIMALS, AND DISCIPLINE Next day Tony left Anna's garden early in the morning. He ambled along the smooth motor road, singing and begging whenever he found someone to beg from. On each side of the road were black posts with white caps on them, glistening in the sun, polished to shine. Snow-white oxen passed, driven by farmers. In vineyards grapevines climbed and twisted about old trees. In nearly every archway a baby, a goat, or a donkey stood and stared at Tony as he passed. Women and children with large baskets or bundles on their heads trudged by. Tiny donkeys carried mountainous loads on their backs.
- 81. ALONG THE ROAD, NEAR NAPLES Occasionally, there would be an automobile. Tony liked the little cars named Balilla, after the Boys' Group. They are the smallest Italian cars
- 82. made. ALONG THE ROAD Tony bought chestnuts and munched them. Chestnuts often take the place of bread among the poor people. Toward the end of day Tony began to miss Tina. He had seldom been separated from her for such a long time. So he returned to Anna's house. He hoped that Anna had not forgotten to leave his dinner. He also hoped that her father would not forget to tell more stories tonight. This was a
- 83. pleasant life. But, of course, Tony was too wise to think that he could go on living like this forever. Guido might find him. Or Anna's father might discover him. Yet if he ran off with Tina he would be a thief like Guido! No, that would never, never do! Oh, how difficult it all was! But upon arriving at Tina's tent he forgot his troubles, for he found there a neatly covered dish. It contained macaroni, meat, and salad. An ideal meal for an Italian boy! Also, Anna had really left some torrone on the plate. Tony sighed with pleasure and began to eat. Good little Anna! All day the little girl had been thinking of the beggar boy. However, she had kept her adventure a secret. But, oh, Tony, beware! Anna is very young, and it is difficult for small children to keep secrets. Especially, when secrets are as interesting as you are! This evening the weather was cooler. The moon did not shine. When Tony finished his dinner, he slipped under the window as he had done before. He hoped Anna's father would tell more stories of Italy. Presently, he saw the family enter the room. They had dined. The mother took up her sewing. The father settled himself in his chair with a book. Anna, with her dog, nestled down in his lap. Tony knew that now more stories were coming. He leaned against the side of the house.
- 84. FLORENCE AND THE ARNO RIVER He closed his eyes contentedly and listened.
- 85. PIAZZA DELLA SIGNORIA: FLORENCE It is early, said Anna's father. We shall have a long time to read tonight. Shall we hear more about the cities of Italy?
- 86. 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