Matching Strings With Mismatches

Question

[Krisr]

I am using perl to match short nucleotide sequences against fasta sequences...

(GeneFasta =~ /$searchSeq/g)

I would like to perform this match, but allow for a mismatch in the search. Does anyone know if, and how, perl may accomplish this?

[Giovanni M Dall'Olio]

this is a bad idea. Why don't you use a short reads aligner instead?

Answer 1

[Jeremy Leipzig]

The Bio::Grep module is pretty good as it provides a common interface for you to interact with several different fuzzy matchers, my favorite being Vmatch

Answer 2

[Aaronquinlan]

agrep (i.e., approximate grep) is a nice tool for this sort of thing. it's not a standard LINUX tool, but it is a good one.

Here's one implementation: ftp://ftp.cs.arizona.edu/agrep/

from the README at the above URL:

" ...for example, "agrep -2 homogenos foo" will find homogeneous as well as any other word that can be obtained from homogenos with at most 2 substitutions, insertions, or deletions. "

[Eric Normandeau]

Thanks. I'm impressed by the quality of this tool.

[Aaronquinlan]

Yeah, believe it not, 3 years ago I hacked it briefly as a short-read aligner.

Answer 3

[Rm]

You are looking for a fuzzy pattern matching program, try perl module String::Approx:

"Perl extension for approximate matching (fuzzy matching)"

For fuzzy pattern matching excercise and scripts go through VCU bioinformatics notes on pattern matching

[Jeremy Leipzig]

I've had some issues with that module - both false positives and misses.

Answer 4

[Alastair Kerr]

Just assigning a regexp to a scalar will not work in perl for sub-sequence pattern matches e.g.

$searchSeq = "AAA[TA]";

Instead you need to use quote regular expression (qr) operator

$searchSeq = qr/AAA[TA]/;