Hi all,
I'm trying to map a primer pair to a reference using bowtie2. One primer maps as expected, but the other just won't, no matter what I try. A pseudo-alignment should look like:
AGAGTTTGATCATGGCTCAG
||| ||||||| ||| ||||
GTAGGCGGCAAGAATTTGATCTTGGTTCAGATTGAACGCTGGCGGCGTGGATGAGGCAT
So there are 3 mismatches, a maximum run of 7, and length of 20. Setting the mismatch penalty (--mp) to 5 should give a score of -15 (I think). There are no Q scores as everything is fasta. The seed length (-L) option is set to 6 and the --score-min option to L,0,-1.2 (which should give a minimum score of -24). As far as I can tell, bowtie2 should return this alignment but it just won't. If I ramp the sensitivity right up, other, nonsensical alignments are called but this obvious one isn't.
Can anyone tell me what I'm doing wrong with this? I would, as ever, be eternally grateful!
Note: I'm aware that bowtie2 isn't necessarily the right tool for this job, but I need a highly scaleable short read mapper that outputs SAM format and allows gaps (so bowtie 1 is out).
bwa mem you tried? for the same