EvanTarbell
commented
5 years ago For motif analysis, I'd actually recommend using the summits defined in the Name_summits.bed file. We showed in our paper that these sites are the most likely locations of TF binding, as defined by the enrichment in PhastCons scores (vertebrate conservation - a hallmark, though no requirement, of TF binding). In fact, any time you're doing motif analysis with ATAC-seq, you should use the summits, even with other algorithms (ie MACS2).
My purpose is to use the open chromatin regions to do MEME analysis to get the potential motif in the open chromatin regions, in this way, which column I should use in the Name_peaks.gappedPeak ? Is the peak start and peak end columns, or the Open state start and end columns? Thanks!