Bowtie Mapping Wrong Strand

[Dario1984]

Hello,

I ran bowtie 0.12.7 on a FASTA file of microarray probe sequences using the hg19 index as supplied on the website, and I noticed from manually inputting into BLAT a couple of sequences that sequences mapping to the negative strand map to + strand with BLAT. For the sequences mapping to + strand with bowtie, they map to the + strand with BLAT.

Can someone check this ? How could it be

bowtie -f -v 0 -m 1 --un probesNotMapping.txt hg19 probeSeqs.txt probesMapped.txt

e.g.

Bowtie output :
- chr16 54901845 AGTATTAAACAAGAATGCACGTAAAGTGTGCAATGCATATAGCAGACAGTCAAGAAATGGTGT

BLAT output :
100.0% 16 + 54901846 54901908 63

For reproducibility, the entire sequences file is here.

[lpagie]

bowtie: reads aligned to reverse strand are returned in reverse complement

just now I thought I had run into the same problem. I was fooled by the fact that the output of bowtie contains the read sequences corresponding to the forward strand. So, when the flag in the SAM output of bowtie places the read on the reverse strand, the reported read sequence will now align on the forward strand.
Check your fasta file you used for input and my guess is that the reads mapped to the reverse strand by bowtie are reverse complement of the reads in the bowtie output.
The bowtie manual (http://bowtie-bio.sourceforge.net/bo...tml#sam-output) indeed specifies that reads in SAM output are reverse-complemented in aligned to the reverse strand.

Ludo