Looks like you got something working! I think you aren't getting quite what you want here. The minOverlap parameter to dba.peakset only is applied if you are adding a consensus peakset; you can't use it and Retrieve the peakset in a single step. The easiest way to get all the peaks that are in at least four peaksets is as follows: > bedfilelist2 = dba(sampleSheet="bedfilelist2.csv",$B!D(B) # read in samplesheet > bedfilelist2 = dba(bedfilelist2, minOverlap=4) # re-compute binding > bedpeakset = dba.peakset(bedfilelist2, bRetrieve=T) Then bedpeakset is a GRanges object with the overlapping peaks, including the ranges and seqnames (chromosomes). Cheers- Rory From: $B9bJ8D6(B-$BCf;3Bg3X(B <gwc19880401@163.com<mailto:gwc19880401@163.com>> Date: Fri, 14 Mar 2014 11:24:11 +0800 To: Rory Stark <rory.stark@cruk.cam.ac.uk<mailto:rory.stark@cruk.cam.ac.uk>> Subject: Have I extracted the right data with DiffBind? Dear Stark, I have 17 peaksets wiht four examples in the mail attachments. I want to extract ranges shared by at least 4 of the 17 peaksets. And then I run the following code under "DiffBind" package: >bedfilelist2=dba(sampleSheet="bedfilelist2.csv",attributes=DBA_ID,con fig="config.csv",bAddCallerConsensus=FALSE, bRemoveM=TRUE, bRemoveRandom=TRUE,bCorPlot=FALSE) >bedpeakset=dba.peakset(bedfilelist2,minOverlap=4,bRetrieve=TRUE) >write.table(bedpeakset@ranges,"overlapdata.txt",sep="\t<mailto:bedpea kset@ranges,="">") >write.table(bedpeakset@seqnames,"seqnames.txt",sep="\t<mailto:bedpeak set@seqnames,="">") The "overlapdata.txt" file provided me of range information and the "seqnames.txt" file provided me of chromosome information of the ranges, is it right? Thanks for your help! Cheers, Gao WenChao Sun Yat-sen University [[alternative HTML version deleted]]