Genomic Regulation Technical Guide

Table of Contents

Letter from the Editor
Index of Experts
Q1: Which histone modification-mapping techniques do you use, and why?
Q2: Which genome-scale methylation mapping techniques do you use, and why?
Q3: Which method do you use to identify and validate microRNA targets?
Q4: What measures do you take to ensure reproducibility in your functional analyses of genomic regulation?
Q5: When using high-throughput sequencing, how do you balance coverage versus cost for any given experiment?
Q6: What are your protocols for data storage and sharing?
Genomic Regulation Grants
List of Resources

Letter from the Editor

Histone modifications, differential methylation, microRNAs — all three work to regulate the genome’s content in their own ways. Whether repressing the expression of certain genes or physically blocking interactions between them, miRNAs and epigenetic marks are forces to be reckoned with. For those researchers who are heeding the call of the genomic regulators, it can be work enough just to stay on top of the current technologies, let alone apply them to answer research questions related to epigenetic modifications and miRNA-mediated expression.

For that, Genome Technology has again called on the experts to resolve your technical, planning phase quandaries. For which applications are arrays better suited to an experiment than sequencing? When using high-throughput sequencing for genome-wide methylation-mapping, what’s the best targeted capture approach and optimal depth-of-coverage? Is that the same for Arabidopsis and Drosophila?

In the pages that follow, academic researchers and core lab directors share tips for getting at genomic regulation with ease and precision. Still need more information on miRNAs and epigenetics? Be sure to consult the additional resources at the end of this guide for the most recent methods papers and Web tools in the field.

— Tracy Vence

Index of Experts

Many thanks to our experts for taking the time to contribute to this technical guide, which would not be possible without them.

Wei Wang
Cornell University
Life Sciences Core
Laboratories Center

Peng Jin
Emory University
School of Medicine

Wei Li
Baylor College of Medicine

Q1: Which histone modification mapping techniques do you use, and why?


We utilize histone ChIP-seq approach, which could provide us the best coverage so far.

— Peng Jin

We normally use ChIP-seq because it is the most popular and cost-effective technology to get the genome-wide histone modification data.

— Wei Li

Q2: Which genome-scale methylation mapping techniques do you use, and why?


We have been using the microarray-based HELP — HpaII tiny fragment enrichment by ligation-mediated PCR — assay for genome-wide DNA methylation screening. Basically, DNA methylation-dependent restriction digestion patterns are characterized on high-density microarrays to infer the methylation state of the restriction sites. Obviously, this assay interrogates only a fraction of all the potential DNA methylation sites, and it also tends to be susceptible to technical variations in the sample processing, but it can quickly screen large numbers of samples at low cost and survey sites genome-wide. For validation of the methylation results from the HELP assay, we apply the Sequenom MassArray EpiTYPER assay on small numbers of selected sites. The Sequenom assay is also high-throughput and cost-effective on a large number of samples.

— Wei Wang

We are developing our own approaches right now, given the problems with commonly used approaches. For example, bisulfite sequencing could not distinguish 5mC from 5-hmC, while MeDIP [methylated DNA immunoprecipitation] could only immunoprecipitate the genomic regions with dense 5mC.

— Peng Jin

We use either whole-genome bisulfite sequencing or reduced representation bisulfite sequencing — RRBS — to profile 5-hmC at single-nucleotide resolution. The unmethylated cytosine is converted to uracil during the bisulfite treatment and sequenced as thymine after PCR amplification, while the methylated cytosine remains unchanged. The methylation ratio is the proportion of remaining cytosines in all the sequencing reads. At current sequencing costs, the former [approach, whole-genome bisulfite sequencing] is still very expensive, while the latter [RRBS] provides an accurate methylation ratio estimate for the genomic regions of interest in a costeffective manner. RRBS employs restriction enzyme digestion targeting CCGG, thus focuses on hotspots of epigenetic regulation, such as promoters and CpG islands. By concentrating on a small portion of the genome, RRBS could yield much higher sequencing depth than a whole-genome shotgun approach.

— Wei Li

Q3: Which method do you use to identify and validate microRNA targets?

We are combining both bioinformatic and proteomic approaches. In general, we utilize multiple prediction programs for initial analyses. We also utilize the SILAC [stable isotope labeling by or with amino acids in cell culture] approach to perform proteomic analyses [in order] to identify the mRNA targets of any given miRNA.

— Peng Jin

Q4: What measures do you take to ensure reproducibility in your functional analyses of genomic regulation?

To investigate the regulation of gene expression by the epigenetic changes in DNA methylation state, both quantities need to be measured in the same sample to look for the correlation between them among multiple samples. For example, on cancer and normal samples, we can measure the methylation profile by HELP assay on DNA samples and gene expression pattern by expression microarray on corresponding RNA samples. Therefore, reproducibility of both assays will contribute to the overall reproducibility of this functional analysis of genomic regulation. In my experience, the biggest source of variation is the sample quality — including both purity and integrity. Stringent quality control needs to be applied on DNA and RNA samples to ensure consistent genomic assay results and good reproducibility. Excluding outlier samples from the genomic data set also improves the overall data quality. Therefore, a larger sample size — i.e. more biological replicates — is always more desirable. To minimize technical variation in sample processing — due to changes in reagents, personnel, and protocol — the whole study had better be completed in a short period of time, although that is not always feasible for large projects.

— Wei Wang

In general, we utilize multiple biological replicates and technical replicates to ensure the reproducibility of our functional analyses.

— Peng Jin

We check the reproducibility according to NIH ENCOD E and Roadmap Epigenome standards (i.e. for DNA methylation and RNA-seq, the Pearson correlation coefficient needs to be greater than 0.9 between replicates; for ChIP-seq, 80 percent of the top 40 percent of targets from one replicate need to lay within the list from the other replicate and vice versa).

— Wei Li

Q5: When using high-throughput sequencing, how do you balance coverage versus cost for any given experiment?

Coverage needs to meet the minimum requirement for any given high-performance sequencing experiment to obtain reliable analysis result to achieve the specific goal of study. Otherwise, inconclusive or compromised results due to insufficient coverage can actually waste the expense and effort of the study. Under a fixed budget, the sweet spot between the number of samples and the persample coverage needs to be determined to answer the particular biological questions. A pilot study, in silico simulation, and the costless literature review will be helpful in the stage of study design. My inclination is to run fewer samples to guarantee sufficient coverage, and then expand the study to include more samples when more funding becomes available. There are different ways to reduce the cost, [such as] for example, multiplexing samples in the same sequencing run to accurately reach the desirable coverage level, [as does] running biological replicate samples instead of technical replicates.

— Wei Wang

It will depend on the type of assays we are performing. For a resequencing project, we would need to get enough coverage of the interest regions. For a ChIPseq project, we typically need to get enough reads to map the peaks with statistical significance. With the [falling] cost of high-throughput sequencing, sufficient coverage would be more important for consideration.

— Peng Jin

Q6: What are your protocols for data storage and sharing?

Our next-gen sequencers are connected to a small local cluster with many hard drives for temporary network storage of the raw sequencing data files produced. After primary and secondary data analysis, deliverable sequence read data files are transferred to a file server connected to a local high performance computer cluster. The raw sequencing data files are archived to tapes and removed from the small cluster for temporary storage after certain period of time. Customers are promoted to download the read files from the file server via secure FTP links sent in the e-mail notification after the sequencing run. By this means, customers have the freedom to share their read data with any collaborator by forwarding the secure links, and they can have quick access to the read data when they use the local high-performance cluster for data analysis. This is also a cost-effective solution for sequencing read file distribution.

— Wei Wang

We are currently utilizing the server at our department for storage, mainly due to the availability. We would like to utilize cloud [computing] for future data storage and sharing.

— Peng Jin

We have, in total, [more than] 50 terabytes of highspeed disk storage, which is located in a dedicated server room and maintained by a senior systems administrator. Data are backed up daily and mirrored to a similar disk storage system in an off-site, secured data center. After rigorous verifications, all the raw and processed data [are converted to their] standard formats, with the proper metadata, and will be deposited in the NCBI Gene Expression Omnibus and Short Read Archive, following established procedures in my lab. We plan to release the data after publication, or one year after data generation regardless of the publication status. In order to provide a uniform platform to facilitate the sharing and comparison of our data, we will establish annual data freezes [in which we create] a snapshot of all data sets that have been made available by the freeze date.

— Wei Li

Genomic Regulation Grants

Organization: National Institutes of Health, National Institute on Drug Abuse
Award: Size and duration will vary according to the nature and scope of the proposed research.
Details: This grant will support research aimed at functional genetics, epigenetics, and non-coding RNAs in drug addiction. The NIH encourages basic genomics research into the fundamental biological mechanisms underpinning addictive processes, including the functional validation of candidate genes and the elucidation of the molecular pathways and processes they involve.
Contact: Scientific/Research, John Satterlee (satterleej@nida.nih.gov); Financial/Grants Management, Deborah Wertz (dwertz@nida.nih.gov)

Organization: National Institutes of Health, National Cancer Institute
Award: Size and duration will vary according to the nature and scope of the proposed research.
Details: The National Cancer Institute intends to support projects that aim to evaluate methylation profiles, histone modifications, and microRNAs associated with the risk of developing cancer in different populations.
Contact: Scientific/Research, Mukesh Verma (vermam@mail.nih.gov); Financial/Grants Management, Crystal Wolfrey (wolfreyc@mail.nih.gov)

Organization: National Science Foundation
Award: Size and duration will vary according to the nature and scope of the proposed research. Awards funded in FY 2010 ranged from $634,846 for five years to $9,946,315 for four years.
Details: NSF intends to support plant genomics projects that aim to address major unanswered questions in plant biology on a genome-wide scale, and is accepting proposals at all scales — from single-investigator projects through multi-institution projects.
Contact: Diane Jofuku Okamuro (dbipgr@ nsf.gov)

List of Resources

Sometimes you need to know more. Here are more sources that may help you out.

Publications

For as many recognized mechanisms of genomic regulation that exist, there are at least twice as many approaches one can take to study each. More still are the options for bioinformatics analysis from which to choose. Here’s a selection of recent methods papers, standby Web tools, and must attend meetings in the field.

Publications

Alexiou P, Manolis M, Hatzigeorgiou AG. (2011). Online resources for microRNA analysis. Journal of Nucleic Acids Investigation. Epub:doi 10.4081/jnai.2011.e4.

Bussotti G, Raineri E, Erb I, Zytnicki M, Wilm A, Beaudoing E, Bucher P, Notredame C. (2011). BlastR — fast and accurate database searches for non-coding RNAs. Nucleic Acids Research. Epub: doi 10.1093/nar/gkr335.

Chen L, Wu G, Ji H. (2011). hmChIP: a database and Web server for exploring publicly available human and mouse ChIPseq and ChIP-chip data. Bioinformatics. 27 (10): 1447-1448.

Chen Y, Meyer CA, Liu T, Li Wei, Liu JS, Liu XS. (2011). MM-ChIP enables integrative analysis of cross-platform and between-laboratory ChIP-chip or ChIP-seq data. Genome Biology. 12: R11.

Deorowicz S, Grabowski S. (2011). Compression of DNA sequence reads in FASTQ format. Bioinformatics. 27(6): 860-862.

Elefant N, Berger A, Shein H, Hofreee M, Margalit H, Altuvia Y. (2011). RepTar: a database of predicted cellular targets of host and viral miRNAS. Nucleic Acids Research. 39 (Suppl1): 188-194.

Fejes AP, Khodabakhshi AH, Birol I, Jones SJ. (2011). Human variation database: an open-source database template for genomic discovery. Bioinformatics. 27(8): 1155-1156.

Francesconi M, Jelier R, Lehner B. (2011). Integrated genome-scale prediction of detrimental mutations in transcription networks. PLoS Genetics. 7(5): e1002077.

Fritz MH , Leinonen R, Cochrane G, Birney E. (2011). Efficient storage of highthroughput DNA sequencing data using reference-based compression. Genome Research. 21: 734-740.

Gu J, Smith ZD, Bock C, Boyle P, Gnirke A, Meissner A. (2011). Preparation of reduced representation bisulfite sequencing libraries for genome-scale DNA methylation profiling. Nature Protocols. 6: 468-481.

Hummel M, Bonnin S, Lowy E, Roma G. (2011). TE QC: an R package for quality control intarget capture experiments. Bioinformatics. 27(9): 1316-1317.

Krueger F, Andrews SR. (2011). Bismark: a flexible aligner and methylation caller for bisulfite-seq applications. Bioinformatics. 27(11): 1571-1572.

Lutsik P, Feuerbach L, Arand J, Lengauer T, Walter J, Bock C. (2011). BiQ Analyzer HT: locus-specific analysis of DNA methylation by high-throughput bisulfite sequencing. Nucleic Acids Research. Epub: doi 0.1093/nar/gkr312.

Muiño JM, Hoogstraat M, van Ham RC, van Dijk AD. (2011). PRI-CAT: a web-tool for the analysis, storage and visualization of plant ChIP-seq experiments. Nucleic Acids Research. Epub: doi 10.1093/nar/gkr373.

Pardo CE, Carr IM, Hoffman CJ, Darst RP, Markham AF, Bonthrom DT, Kladde MP. (2011). MethylViewer: computational analysis and editing for bisulfite sequencing and methyltransferase accessibility protocol for individual templates (MAPit) projects. Nucleic Acids Research. 39 (1): e5.

Qin J, Li MJ, Wang P, Zhang MQ, Wang J. (2011). ChIP-Array: combinatory analysis of ChIP-seq/chip and microarray gene expression data to discover direct/indirect targets of a transcription factor. Nucleic Acids Research. Epub: doi 10.1093/nar/gkr332.

Shankaranarayanan P, Mendoza-Parra MA, Walia M, Wang L, Li N, Trindade LM, Gronemeyer H. (2011). Single-tube linear DNA amplification (LinDA) for robust ChIPseq. Nature Methods. Epub: doi 10.1038/nmeth.1626.

Shen Y, Song R, Pe’er I. (2011). Coverage tradeoffs and power estimation in the design of whole-genome sequencing experiments for detecting association. Bioionformatics. Epub: doi 10.1093/bioinformatics/ btr305.

Vavouri T, Lehner B. (2011). Chromatin organization in sperm may be the major functional consequence of base composition variation in the human genome. PLoS Genetics. 7(4): e1002036.

Wang C, Zhang D. (2011). A novel compression tool for efficient storage of genome resequencing data. Nucleic Acids Research. 39(7): e45.

Zisoulis DG, Yeo GW, Pasquinelli AE. (2011). Comprehensive identification of miRNA target sites in live animals. Methods in Molecular Biology. 732: 169-185.

Websites

ChromDB
http://www.chromdb.org/

MethDB
http://www.methdb.de/

miRBase
http://www.mirbase.org/

miRDB
http://mirdb.org/miRDB/

miRNA Target Database
http://www.ncrna.org/KnowledgeBase/linkdatabase/mirna_target_database

NCBI Gene Expression Omnibus
http://www.ncbi.nlm.nih.gov/geo/

NCBI Short Read Archive
http://www.ncbi.nlm.nih.gov/Traces/sra

NHGRI Histone Sequence Database
http://research.nhgri.nih.gov/histones/

PubMeth
http://www.pubmeth.org/

TargetScan
http://genes.mit.edu/targetscan/index.html

Conferences

Epigenetics: Mechanisms, Development, and Disease
Gordon Research Conferences
Aug 7-12, 2011
Easton, Mass.

Epigenetics Europe
Select Biosciences
Sep 8-9, 2011
Munich, Germany

RNAi & miRNA Europe
Select Biosciences
Sep 8-9, 2011
Munich, Germany

Epigenomics of Common Diseases
Wellcome Trust
Sep 13-16, 2011
Hinxton, UK

EM BO Workshop: Histone Variants & Genome Regulation
European Molecular Biology Organization
Oct 12-14, 2011
Strasbourg, France

IN SERM Workshop: High-Throughput Approaches in Epigenomics
Institut National de la Santé et de la
Recherche Médicale
Oct 10-12, 2011
Bordeaux, France

INSERM Workshop: Bioinformatics
Approaches to Decipher Genome Regulation
Institut National de la Santé et de la Recherche Médicale
Oct 12-14, 2011
Bordeaux, France

MicroRNAs Europe 2011
GeneExpression Systems
Nov 1-2, 2011
Cambridge, UK

Genome Informatics
Cold Spring Harbor Laboratory, Wellcome Trust
Nov 2-5, 2011
Cold Spring Harbor, NY

X CRG Annual Symposium: Computational
Biology of Molecular Sequences
Centre for Genomic Regulation
Nov 10-11, 2011
Barcelona

Next-Generation Sequencing
Congress Europe
Oxford Global Conferences
Nov 14-15, 2011
London

EuroEpiStem: European Epigenomics & Stem Cells
GeneExpression Systems
Nov 21-22, 2011
Paris

RNAi Asia
Select Biosciences
Nov 22, 2011
Singapore

Chromatin: Structure & Function Abcam
Dec 5-8, 2011
Aruba

Epigenomics
Keystone Symposia
Jan 12-22, 2012
Keystone, Colo.

Gene Silencing by Small RNAs
Keystone Symposia
Feb 7, 2012
Keystone, Colo.