christophertbrown / bioscripts Goto Github PK

I have been trying to use iRep which seems to use the functionctbBio.mapped.get_reads() and ctbBio.mapped.reads_from_mapping() from this package to parse SAM files. The documentation mentions that the parser expect reads from the same pair to be one after the other in SAM file. Why not.

The problem I have is that I have reads with unmapped mate in my SAM files, and I suspect it could be the source of (silent) incoherent coverage results I get when I try to use iRep.

The parser in ctbBio tries to take this into account in its own way:
https://github.com/christophertbrown/bioscripts/blob/master/ctbBio/mapped.py#L122-L150
In particular:

 if int(line[1]) <= 20: # is this from a single read?

I just checked, and the FLAGs (line[1] in the code above) of my single reads are way above 20. In fact, I do not understand where this 20 comes from, given the specifications of the SAM format https://broadinstitute.github.io/picard/explain-flags.html. For instance, I have single reads with FLAG 137, which means "read paired", "mate unmapped" and "second in pair".

Could you give a little more details about how ctbBio.mapped handles unmapped mates in the SAM file?