This is a test version of Biostars. For the public version, visit https://www.biostars.org.
bwa XA field: Why are position integers signed?

Hello All, An example syntax of the bwa XA field is:

XA:Z:22,-32166766,35M,0;

More generically, the bwa manual http://bio-bwa.sourceforge.net/bwa.shtml describes the following:

Alternative hits; format: (chr,pos,CIGAR,NM;)*

For the XA field.

The compulsory POS (alignment position / 4) field in sam format is an unsigned integer. Why is the pos field in the XA format signed? I'm guessing it is relative to something? Is it relative to the value in the POS field?

For example, if POS = 5000, and we have an alternative alignment XA:Z:_,-2000,_,_; where _ could be anything, is this saying that the alternative alignment starts at 3000?

What if this runs across chromosomes? I.e REF = 14, being, alignment to chr14, but the XA field is XA:Z:22,-3200000,_,_; Would we simply compute POS - 3200000 to find the alternative alignment position value?

Thanks,

Izaak

bwa bwa aln sam
• 3,783 views • link
written by izaak.coleman1 • izaak.coleman1

1 answer

Apologies, an equivalent question and answer has been posted here. I should note that the answer given in the link is based on bwa source code. If anyone can find a documentation (rather than source code) based answer, please post it here - reason being, if individuals do not understand the code, they will not understand the evidence that determines the meaning of signage.

:D

• 0 views • link
written by izaak.coleman1 • izaak.coleman1

Log in to answer this question.

More posts like this