How does diffbind extract the original coordinates of each difference peak

Question

[441559287]

After using dba. analyze() to calculate the difference, use dba. report() to extract the difference peak, but the difference peak coordinates have been merged. How to output the real peak coordinates corresponding to each difference peak in different sample species?

> library(DiffBind)

> dbObj_S3_S2 <-dba(sampleSheet="D5_D3_Linux.csv")
> dbObj_S3_S2 <- dba.count(dbObj_S3_S2 , bUseSummarizeOverlaps=TRUE)
> dbObj_S3_S2 <- dba.normalize(dbObj_S3_S2)
> dbObj_S3_S2 <- dba.contrast(dbObj_S3_S2,reorderMeta=list(Condition="S2"),minMembers = 2)
> dbObj_S3_S2 <- dba.analyze(dbObj_S3_S2, method=DBA_ALL_METHODS)
> comp1.deseq2_S3_S2 <- dba.report(dbObj_S3_S2, method=DBA_DESEQ2,contrast=1,th=1)

Answer 1

[Rory Stark]

I'm not sure what you mean by "real" peak coordinates.

Do you mean the overlapping peak calls in each sample that contributed to the consensus peak? In that case there can be multiple "real" peak coordinates for each consensus peak. You can identify these by overlapping the consensus peaks returned by dba.report() with each individual peakset. The easiest way to do that is to load the individual peaksets as GRanges objects (as is the returned report).