Ok, thanks - I misread the manual, I just thought that was a way to filter out reads falling over certain regions. After reading the alignments I use summarizeOverlaps to obtain counts with the desired options. I decided to use scanBamParam(which=..) because loading bams and counting overlaps is using a lot of RAM for big bams and I found online various suggestions to use the ScanBamParam to load one chr at a time...but as long as I would like to be general I decided to extract all the seqlevels in the GRanges object that I'm using and then I would like to iterate over those extracting the reads that fall on different seqlevels. I noticed that the counts where different with this approach with respect to doing all the work together. I guess that I have to find another way to split my alignments...but I'd really like to avoid something like: http://comments.gmane.org/gmane.comp.lang.r.sequencing/755 Reducing my features would not work as long as I need to keep the original number of the reads...I guess I will have to "flatten" all the regions falling on a single chr of my Granges in a way that avoid overlapping reads, that should do the trick (I hope...)...maybe I could extract the smaller-bigger coords for a given chr and obtain chr-based Granges from there.
Thank you very much for your help, E. _______________________________________________ Bioc-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/bioc-devel
