search

tseemann / prokka

:zap: :aquarius: Rapid prokaryotic genome annotation
843 stars 226 forks source link

Setting minimal coverage of blast against sprot #447

Closed dyogoborst closed 4 years ago

dyogoborst commented 4 years ago

Hi, I noticed there isn't a minimal coverage parameter in the primary blast against the swissprot database. Is there an easy way for me to specify this for myself? I would like the output to be more accurate so that I can align the remaining 'hypothetical proteins' with a subset of the nr databse of NCBI in a local blast.

To give you an example, I annotated a contig where a lukF-PV gene was localized but it was annotated as lukDv with a similarity of 83%, my theory is that if I would raise the MINCOV to a minimum of 90% , this particular gene would be assigned as 'hypothetical protein' and I could still identify it as lukF-PV with an local blast.

Thanks in advance!

tseemann commented 4 years ago

Are you using this parameter?

  --coverage [n.n]   Minimum coverage on query protein (default '80')

What version of Prokka are you using?

dyogoborst commented 4 years ago

Hi, Thanks for your quick reply, I didn't see this parameter and will test this in the coming week. I'm currently using Prokka v1.13.4 in my conda environment.

tseemann commented 4 years ago

You should upgrade to 1.14.5 which I think is in conda.