how to merge sam files from mapping the same fq to different reference

[pengchy]

Dear all,

I want to know if there some utilities to merge sam files produced from mapping the same reads file to the different reference sequences. The necessity occurred when the reference sequence so large that the mapper can not process.

Thank you in advance.

pengchy

[maubp]

That is tricky.

What would you expect to happen when a read is mapped nicely to both reference files?

Are you using paired end data? Consider the case where each read has mapped nicely to a different reference.

P.S. How long are your reference sequences? I'm trying to encourage the SAM/BAM community to think about supporting very large chromosomes as in plants.

[pengchy]

thank maubp for your reply.

Because the reads can only be produced from one place, so it is expected giving only one randomly. We care about the unique mapped reads or the randomly one place for the multimapped reads. So, if the both ends uniquely mapped to one place, it will be not a matter. For the multi hit pe-reads, the proper paired position, or randomly one if there are many, will be expected.

This is direction will be a question for the future large genome that current aligner can not tackle.

[maubp]

Would you also expect some read should map across the break point? Or can you choose where to break the long reference sequence (e.g. a large region NNNN from scaffolding, or an area annotated with no genes).

[I am assuming you have had to break up a long chromosome - things are easier if you just have divided up the chromosomes into one FASTA reference file per chromosome]