Dear Rory Stark,

Thanks for DiffBind package for analyzing ChIP-seq dataset that has multiple replicates. I encountered two issues when trying to add a derived consensus peak to an existing DBA object.

1. dba.overlap(mydba, mode=DBA_OLAP_RATE, mask=mydba$masks$tissue)

      Error: No peak overlap information available

2. my.new.consensus.peakset = dba.peakset(my.dba, consensus = -DBA_REPLICATE, minOverlap=3)

   Error in dba.peakset(my.dba, consensus = -DBA_REPLICATE,  :
   DBA object is already formed from a consensus peakset!

Is there any other parameters that I should include? I basically copied from "tamoxifen" example code here.

Thank you!

Yuhong

I'd need to see how you constructed your DBA object (my.dba). Did you use dba,peakset() with consensus=TRUE? Basically if you added a consensus peakset yourself, and didn't have DiffBind computed one from a set of underlying peak sets, there is no way to know which consensus peaks are associated with which of the original peaksets (hence "No peak overlap information available).

Perhaps you could share your construction code and/or your DBA object my.dba with me so I can have a look? You can email the object to me.

Cheers-

Rory