I'm trying to map 50PE Illumina reads to a couple reference genomes, and then sort out later which ones are unambiguously from one species or another. I'm trying to use bowtie2 to do the mapping, with the following options:
bowtie2 -p 8 --all --no-mixed --local
And when it runs, it happily reports that I have plenty of reads that align concordantly many times:
34669986 reads; of these:
34669986 (100.00%) were paired; of these:
4165729 (12.02%) aligned concordantly 0 times
23793250 (68.63%) aligned concordantly exactly 1 time
6711007 (19.36%) aligned concordantly >1 times
However, when I check the actual sam file, I find that there are no reads where the "secondary alignment" flag bit is set. In particular,
$ awk 'and(256, $2)' accepted_hits_unsorted.sam
returns nothing. Is the --all flag broken? Am I being boneheaded in an obvious way?
Thanks for the help!
bowtie2 -p 8 --all --no-mixed --local
And when it runs, it happily reports that I have plenty of reads that align concordantly many times:
34669986 reads; of these:
34669986 (100.00%) were paired; of these:
4165729 (12.02%) aligned concordantly 0 times
23793250 (68.63%) aligned concordantly exactly 1 time
6711007 (19.36%) aligned concordantly >1 times
However, when I check the actual sam file, I find that there are no reads where the "secondary alignment" flag bit is set. In particular,
$ awk 'and(256, $2)' accepted_hits_unsorted.sam
returns nothing. Is the --all flag broken? Am I being boneheaded in an obvious way?
Thanks for the help!