A brief review of recent ChIP-Seq tools literature

[ECO]

Seems I can't go a week without finding a paper describing a new ChIP-Seq analytical tool, so I thought it would be helpful to put together a list of a few of the most interesting papers in this area.

For those not familiar with ChIP-Seq, it involves massively parallel sequencing of nucleic acids recovered from a chromatin-immunoprecipitation (ChIP). ChIP uses an antibody to capture protein that has been previously crosslinked to DNA in its native state. Thus, ChIP-Seq allows sequencing of all DNA that a specific protein physically interacts with within the cell.

Courtesy the ChIP-SEQ Wikipedia entry

There are quite a few studies recently which describe either a software tool, and/or an analytical framework for working with this type of data. The difficulty and thus the plethora of approaches is using the read density information to "call peaks" or more precisely, statistically significant ChIP-Seq binding sites.

Recent papers I have found helpful are below, in reverse chronological order. Please accept my apologies in advance if I've missed your favorite tool.


November

• F-Seq: a feature density estimator for high-throughput sequence tags., Duke University. F-Seq Project Homepage

• CisGenome: An integrated software system for analyzing ChIP-chip and ChIP-seq data., Johns Hopkins. CisGenome homepage

October

• ChromaSig: a probabilistic approach to finding common chromatin signatures in the human genome., UC San Diego. ChromaSig homepage

• An HMM approach to genome-wide identification of differential histone modification sites from ChIP-seq data., Genome Institute of Singapore. ChiPDiff Homepage

September

• Model-based Analysis of ChIP-Seq (MACS)., Dana-Farber Cancer Institute and Harvard School of Public Health. MACS Homepage

• A Poisson mixture model to identify changes in RNA polymerase II binding quantity using high-throughput sequencing technology., Indiana University School of Medicine

• Genome-wide identification of in vivo protein-DNA binding sites from ChIP-Seq data., NHLBI, NIH. SISSRs Homepage

August

• Modeling ChIP sequencing in silico with applications., Yale University. ChIP-Seq Simulator Homepage

• Genome-wide analysis of transcription factor binding sites based on ChIP-Seq data., Stanford University Medical Center (woohoo!), QuEST Homepage

• FindPeaks 3.1: a tool for identifying areas of enrichment from massively parallel short-read sequencing technology., BCGSC (First author is our very own apfejes!). Vancouver Short Read Analysis Package Homepage. Version 3.2 is out.

Hopefully those are useful. I'll post up my experiences with some of these tools soon, and would love to hear yours.

[Roald]

What a great list - thanks a lot for posting!

[ECO]

Just when I think I captured them all...another one comes out!

CisGenome

[apfejes]

Hey ECO,

Awesome list, I can't believe how many people there are writing ChIP-Seq software, now.

At any rate, this is a great resource for people just getting into the field. (=

Can I ask that you put a link to the FindPeaks 3.2 homepage, now? I've already got pre-compiled jar files up on the web, so maybe it's time to start pointing people in that direction (=

http://vancouvershortr.sourceforge.net/

Anthony

[foolishbrat]

Hi Eco,

Great list.

BTW, do you happen to know any literature that does the error correction model of tag counts.

[Nix]

Here's our publication, it's finally out:

http://www.biomedcentral.com/1471-2105/9/523/abstract

And the project web site:

http://useq.sourceforge.net/

I'd like to suggest running a bake off using some simulated spike-in data until real spike-in data becomes available. This was critical for evaluating the methods used in the USeq package. Basically add simulated spikes to real input DNA sequencing data then run your favorite software to call peaks. I'm near getting together a stranded dataset that would be a good first start. Would be nice to sweeten the pot though. Maybe we could get Illumina to throw in an iPod to the winner?

[dnewkirk]

Bioconductor is in the process of developing an R package (appropriately labeled chipseq currently) which should be released in the not too distant future. A working version can be had from the developers subversion repository, but again its incomplete and may change over time. Conversion of Data objects is a little bit troublesome, but the mailing list can be browsed for tips on this.

[ewilbanks]

Hi all,

Anyone have updates to this list of programs?

[apfejes]

I'll be updating FindPeaks (Vancouver Short Read Analysis Package) to version 4, shortly.... is there something specific you're looking for?

[ewilbanks]

Cool! Not looking for anything in particular, just trying to make sure I've checked out the different algorithms out there

[Chema]

Hi,

In this paper, you can find other ChIP-seq algorithm. It is implemented in R
http://www.pubmedcentral.nih.gov/art...?artid=2671559

The baisc idea was
1) Extend the reads and use the minimum number of read that overlap a a nucleotide position among both strnad. This is, the number of reads should be supported by the two strands independently. This is to avoid peaks that have a block shape.
2)Normalize data and sample regarding their covarage distribution
3)Use permutation to estimate the FDR control

[ieuanclay]

might want to add PeakSeq too:

http://www.nature.com/nbt/journal/v2.../nbt.1518.html

Not earth-shattering, but i think their advice on normalising to input is really worthwhile.

[ewilbanks]

Two new ones out there:

GLITR: http://web.me.com/kaestnerlab1/GLITR/

and

TIROE: http://www.pnas.org/content/106/13/5187.abstract

[david.lapointe]

Is CHiPSeq from the Wold lab not available any longer. The links seem to go nowhere.

[Nicolas]

Originally Posted by david.lapointe

Is CHiPSeq from the Wold lab not available any longer. The links seem to go nowhere.

Is that what you are looking for?

http://woldlab.caltech.edu/rnaseq/

[tujchl]

A wonderful work!!!
thank you

[alperyilmaz]

Where can I find the results that Nix suggested earlier, same sample sequences tested with different peak finding software?

[Nix]

Here's the thread to the ChIP Seq Challenge http://seqanswers.com/forums/showthread.php?t=1039

[sameet]

Hi All,

We recently completed first chip-seq run in our laboratory. We could find a few good peaks with MACS. What I want to do is find the distribution of the distances of all the genes that are closes to the peaks on their respective 5' and 3' end. Is there a simple way to do it. Naturally I have the genomic coordinates. If anybody has done this before, it will help most helpful else I will try to write a tool to get it done.

regards

[ewilbanks]

You might want to check out the following tool, which has a built in "nearest gene finder". http://havoc.genomecenter.ucdavis.ed...in/chipseq.cgi

Cheers,
Lizzy