Missing "#=GF ID" with "-A" option on nhmmer (& presumably hmmsearch) searches with multiHMMs.

[ppgardne]

Hi Sean,

Just spotted a minor issue. We've used a multi-hmm to search a bunch of genomes with either nhmmer or hmmsearch. When we use "-A" we get some lovely Stockholm alignments, but it'd be handy to have a "#=GF ID" line containing the name of which hmm the alignment was generated from (each hmm has a corresponding unique e.g. "NAME GlmY". e.g.

STOCKHOLM 1.0

=GF ID GlmY

=GS CP007259/3866060-3867261 DE [subseq from] [subseq from] CP007259

... //

We can map each alignment back to each hmm using "tblout" outputs, but it'd be much easier with ID lines. Then we can esl-afetch goodness too.

Some minimal illustration+versions below.

I hope all is well in Boston. I see the US is losing the Olympics. ;-) http://www.medalspercapita.com/

Cheers, @ppgardne & @biobeth

nhmmer :: search a DNA model or alignment against a DNA database

HMMER 3.1b2 (February 2015); http://hmmer.org/

nhmmer -A test.stk --tblout test.tblout sRNA.hmm test.fa > test.nhmmerout

Produces:

STOCKHOLM 1.0

=GS CP007259/3866060-3867261 DE [subseq from] CP007259

... CP007259/3866060-3867261 GAAT... ... //

hmmsearch :: search profile(s) against a sequence database

HMMER 3.1b2 (February 2015); http://hmmer.org/

hmmsearch -A test.stk --tblout test.tblout sRNA.hmm test.fa > test.hmmsearchout

Produces:

STOCKHOLM 1.0

=GS CP007259/3866060-3867261 DE [subseq from] [subseq from] CP007259

... CP007259/3866060-3867261 GAATGC... ... //

[cryptogenomicon]

Done. hmmsearch, jackhmmer, nhmmer, phmmer and hmmpgmd will now put name, accession, description, and author (i.e. "hmmsearch 3.1b3") into an MSA that's saved with -A. This will appear in the next HMMER3 release Real Soon Now. We appreciate your business and we will punish you for the gratuitous anti-American crack at the appropriate time.