Dear devs,

I am working with the latest version of featureCounts program (1.5.2), and especially on RIBO-seq data. I do need to reduce reads to one position. To do so, I use the very useful --read2pos option.

However, I do need an extra processing to adjust the position to the P-site of the ribosome (+12 from the 5' end), and I was wondering me that, does featureCounts has a solution to use --read2pos INT? In my case, INT = 12, and all reads will be reduced to one position, the P-site of the ribosome.

I understand that this implementation will not be trivial as the CIGAR has to be taken in account and the strand is important too, but I do think that this implementation will be very useful for all bioinformaticians / biologists working with RIBO-seq data to avoid an extra-processing step to adjust the counts and still working with featureCounts, which is a very robust and fast program.

Thank you very much for your time and answers.

Glihm

One way to get around this is to change your annotation - subtracting/adding each P-site coordinate by 12 (depending on which strand the P-site is on).

In the meantime, we are working on letting read2pos parameter work with readExtension5 parameter (and readExtension3 parameter) to reduce the read sequence to any single base position.