Find all occurrences of a sequence in a fasta file

[dphansti]

I have a fasta file with 16S sequences from many organisms. I want to find all occurrences of a certain ~20 bp sequence in this fasta file. I could do a simple text search but I would prefer to allow some flexibility in the matches.

For each match I would like the following information
1) fasta entry name
2) postion in the sequence
3) CIGAR string or some other representation of the alignment

A SAM file would be fine. I tried using bowtie2 with "-a" but it never seemed to finish. Through trial and error I found that setting "-k" to 150 worked fine but setting "-k" to 200 did not, indicating to me that there is probably some upper limit to the number of matches per query that it can report.

I am certain that what I want to do is commonly done by many people here on the site. What is the easiest/best way to go about it?

Thanks so much.

[gringer]

Have you tried BLAST or BLAT? Those tools are designed for looking for a low number of sequences in a large database of many different sequences.

[aloliveira]

You can use BLAST program and specify the parameter -W 20 (in that way BLAST will report just the hits with at least 20 pb of similarity). Ahhh, an important think deactivate the low complexity filter (-F F)

Thank's
André

[vamosia]

I would suggest primer_match. It is rather simple to use and rather flexible in its output.

The primers in this specific case would be your 20bp sequence. The program would in turn allow you to manipulate the output displaying position, entry name, or counts

http://edwardslab.bmcb.georgetown.edu/software/primer_match.html

(I am by no means associated with edwards lab, just a frequent user)