ChIP-seq peak calling from replicates

ttnguyen · 07-25-2011, 05:12 AM

Hi,
Are there any advice or software to do peak calling for histone modification ChIP-seq data having two replicates with two corresponding control libraries? Thanks a lot.

emilyjia2000 · 08-07-2011, 09:05 AM

I have similar data, which is technique replicate, I am thinking merge the two replicates before peak calling. any suggestions?

Thanks

mudshark · 08-07-2011, 11:40 PM

the easiest way is to do the peak calling on each data set and retain the common peaks. however, histone modificiations might not necessarily give a peak-like pattern rather extended regions of enrichment. this could be more difficult to analyze.

emilyjia2000 · 08-09-2011, 05:53 PM

Hi, shark,
How do judge if the peaks are common or not between the two replicates? I am asking because very few peaks could be exactly same, some of them could be overlapped more or less between two replicates.
Thanks

mudshark · 08-10-2011, 01:21 AM

that should depend on the nature of your target. for sequence specific transcription factors I would consider common peaks to be located (peak center) within 200 bp. for histone modifications this is more difficult to define.

in case you have similar amount of reads in all your samples you could also do a simple enrichment statistic on windowed data (100 bp windows e.g.). as I already said histone modifications might have extended regions of enrichment and peak calling is not the best way to look for those.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
ChIP-Seq: Picking ChIP-seq peak detectors for analyzing chromatin modification experi	Newsbot!	Literature Watch	1	06-14-2012 04:31 AM
ChIP-Seq: PeakRanger: A cloud-enabled peak caller for ChIP-seq data.	Newsbot!	Literature Watch	0	05-11-2011 03:40 AM
ChIP-Seq: Rapid innovation in ChIP-seq peak-calling algorithms is outdistancing bench	Newsbot!	Literature Watch	0	11-10-2010 02:40 AM
ChIP-Seq: Probabilistic Peak Calling and Controlling False Discovery Rate Estimations	Newsbot!	Literature Watch	0	09-10-2010 02:10 AM
ChIP-Seq: Evaluation of Algorithm Performance in ChIP-Seq Peak Detection.	Newsbot!	Literature Watch	0	07-16-2010 02:00 AM

07-25-2011, 05:12 AM	#1
ttnguyen Member Location: Ireland Join Date: Mar 2010 Posts: 41	ChIP-seq peak calling from replicates Hi, Are there any advice or software to do peak calling for histone modification ChIP-seq data having two replicates with two corresponding control libraries? Thanks a lot.

08-07-2011, 09:05 AM	#2
emilyjia2000 Member Location: usa Join Date: May 2011 Posts: 59	I have similar data, which is technique replicate, I am thinking merge the two replicates before peak calling. any suggestions? Thanks

08-07-2011, 11:40 PM	#3
mudshark Senior Member Location: Munich Join Date: Jan 2009 Posts: 138	the easiest way is to do the peak calling on each data set and retain the common peaks. however, histone modificiations might not necessarily give a peak-like pattern rather extended regions of enrichment. this could be more difficult to analyze.

08-09-2011, 05:53 PM	#4
emilyjia2000 Member Location: usa Join Date: May 2011 Posts: 59	Hi, shark, How do judge if the peaks are common or not between the two replicates? I am asking because very few peaks could be exactly same, some of them could be overlapped more or less between two replicates. Thanks

08-10-2011, 01:21 AM	#5
mudshark Senior Member Location: Munich Join Date: Jan 2009 Posts: 138	that should depend on the nature of your target. for sequence specific transcription factors I would consider common peaks to be located (peak center) within 200 bp. for histone modifications this is more difficult to define. in case you have similar amount of reads in all your samples you could also do a simple enrichment statistic on windowed data (100 bp windows e.g.). as I already said histone modifications might have extended regions of enrichment and peak calling is not the best way to look for those.