bowtie - multiple mapped sequences

[sekta]

Dear Bowtie Users,

I noticed that in my mapping results some uniquely mapped sequences are reported to be mapped multiple times. How come?



I am mapping to miRbase data base. I prepared indexes twice as I did mapping. Each time the same result.

Ps. Mapping options:
sudo ./bowtie -p 1 -f -v 0 -l 15 -a indexes/miRbase samples/CPM15_SignReads.fa map_m_v0.map

Could you please help me and solve the mystery?

[dpryan]

Try using the SAM output format, nothing reads the bowtie format anymore and it lacks too much information to make much sense out of anything.

[sekta]

Dear Dpryan,

thank you for you advice, here I put the same bowtie output, but in SAM format

ania@ania:~/Programy/bowtie-1.0.0$ cat map_m_v0.sam | grep TGTAAACATCCTTGACTGGAAGCT
3356 0 hsa-mir-30e 17 255 24M * 0 0 TGTAAACATCCTTGACTGGAAGCT IIIIIIIIIIIIIIIIIIIIIIII XA:i:0 MD:Z:24 NM:i:0
3534 4 * 0 0 * * 0 0 TGTAAACATCCTTGACTGGAAGCTT IIIIIIIIIIIIIIIIIIIIIIIII XM:i:0
ania@ania:~/Programy/bowtie-1.0.0$

[ecSeq Bioinformatics]

These are two different sequences! Let me write it different, that you can see the difference:
TGTAAACATCCTTGACTGGAAGCT <- your grep call
TGTAAACATCCTTGACTGGAAGCT <- hit 1 (found in your db)
TGTAAACATCCTTGACTGGAAGCTT <- hit 2 (not mapped to your db)

[sekta]

Yes, so why im my results in .map format there is one more alignment for TGTAAACATCCTTGACTGGAAGCT reported?

** 1 at the end of the line in .map format means "one more", there should be 0 as the alignment is unique

[dpryan]

I wouldn't be surprised if the function that outputs the default bowtie format has a bug, since no one uses it anymore.

[sekta]

Thanks Dpryan, probably you are right.