Hi,

I extracted my consensus peakset from the DBA object after counting. I filtered this list to remove outliers and other criteria. I would like to change the consensus peakset in my DBA object now, using my filtered list.

I extracted the consensus peaks like this:

countTable <- dba.peakset(sampleCounts, bRetrieve=T,DataType=DBA_DATA_FRAME)

How do I put it back into the DBA object?

Hi Sarah-

The most straightforward way is to pass in the filtered peaks to dba.count():

> countTable <- dba.peakset(sampleCounts, bRetrieve=T,DataType=DBA_DATA_FRAME)
> filteredPeaks <- do.some.filtering(countTable) #your filtering code
> filteredCounts <- dba.count(sampleCounts, peaks=filteredPeaks)

This will result in the new, filtered set of peaks being used as the the consensus peakset. It will also require that all the reads are re-counted for all the samples, so it is not a "free" operation. If you are using the default read score for your initial call to dba.count(), you would have retrieved normalized read scores (and presumably used them for your filtering). These scores would need to be re-normalized after filtering, which dba.count() will do.

Hope this helps-

Rory

Hi Sarah-

Sorry for the delay in responding. 

Could you let me know which version of the software you are running? Result of sessionInfo() or packageVersion("DiffBind"). This works for me in the most recently released version (2.0).

-Rory