Hi everyone, I am trying to use DiffBind to process an unusually large dataset (101 ATAC-seq samples). Is there a way to construct the dba object using dba() in parallel? The samples take a very long time to load that many BED and BAM files. Looking at the documentation, I don't see the bParallel option for the initial construction of the dba object. Thanks!

Are you using a recent/current version? This function has been optimized for performance at various points from DiffBind_3.4.0 onwards, but it will depend on the numbers of peaks being merged. How big are your peak files (BED)?