Is there a way to set minimum subject coverage for BLAST?

Question

[crystal4]

I have ORFs picked from a metagenomics dataset, and I want to see if any of them match sequences from a database of genes of interest. The ORFs are the query sequences and the database sequences are the subject. The problem is, some of the ORFs are very short, so they will match very well to a small portion of one of the database sequences but that isn't useful information to me; I only want matches where most of the subject is covered by the query. I know there's -qcov_hsp_perc to set a minimum query coverage, but is there a way to also set minimum subject coverage?

Answer 1

[lieven.sterck]

I don't think so from the top of my head (without looking deeply in the manual thus).

What you could do is run a normal blast, ask for the tabular output format (you can customise that one and add the slen value , =subject length) , and run that trough an awk cmdline which compares the subject length hit to it's total length .

Apart from potentially not being present as a parameter, setting to restrictive parameter values in BLAST tends not to be a good idea in most cases. It's better to run it as default as possible and afterwards filter the results to your likings. Can't seem to find the direct link to more on this issue but starting from https://blastedbio.blogspot.com/2015/12/blast-max-target-sequences-bug.html you will get there as well.

Answer 2

[Mensur Dlakic]

You could go for global alignments with FASTA.