0
$\begingroup$

I know how to do a 80-mer window search against a single protein

fasta35 -k 0 -w 80 -a -m 9 -C 20 -Q -3 -E10 -d20 -B tempQuery:0-80 

FASTAbyGI15/378405189 
fasta35 -k 0 -w 80 -a -m 9 -C 20 -Q -3 -E10 -d20 -B tempQuery:1-81 FASTAbyGI15/378405189 
fasta35 -k 0 -w 80 -a -m 9 -C 20 -Q -3 -E10 -d20 -B tempQuery:2-82 FASTAbyGI15/378405189 

....

This will give me the detailed results but I want to do the search against a database in Fasta like this website:

http://www.allermatch.org/allermatch.py/form

Can anyone help me?

$\endgroup$
  • $\begingroup$ Fasta utilizes a k-tuple method to fasten up the search. You could try and set the default ktup parameter to 80 ( you might have to recompile from source to do that). Or you could try brute force and manually chop your input to 80 amino acid windows and call fasta search for each, and then collect all the results and filter out those that are above the threshold. $\endgroup$ – Nandor Poka Apr 4 '15 at 14:55
  • 1
    $\begingroup$ I don't think fasta is the tool you want use, because the speed of a fasta search against a database involves a step where each record in the database is chopped up into shorter substrings, e.g. 3-mers or 4-mers, and that pre-compute step takes awhile. Then, for each search your query sequence is similarly chopped up into a list/table when you submit the job. Your query runs quite fast (yes that is how they named it) compared to the classical hamming distance calculation, because the best hits are targets with the largest number of k-mer matches (that is a slight simplification). $\endgroup$ – mdperry Apr 4 '15 at 14:58
  • 1
    $\begingroup$ Yeah, what @poka.nandor said. I was going to add that classical textbook dynamic programming would be another approach to explore $\endgroup$ – mdperry Apr 4 '15 at 14:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.