Hmmer for protein prediction

Question

[vishalchanda364]

Hello, Regarding using the Hmmer (3.1b1) my requirement is to search amino acid sequences of predicted protein products against an HMM database prokaryotic viral orthologous groups. Which I know can be done using hmmscan.

The user guide says hmmpress has to be performed to prepare a database to run hmmscan on the the test file. But the vog file from the website ( http://dmk-brain.ecn.uiowa.edu/pVOGs/downloads.html ) downloads a tar.gz file and hmmpress asks for a Hmmer3 file. I have no idea about this step and getting confused over it.

Can anyone help me in this reagrds,Please?

[lieven.sterck]

you will for sure have to unpack the tar.gz file you downloaded (it will normally contain a number of other files)

tar -zxvf <download.tar.gz>

should do the trick

[vishalchanda364]

Yes, I had already done that. Thank you.!

[lieven.sterck]

ok, so what was in the un-zipped tar file then? I'm guessing nothing that could be used for resolving your issue?

[vishalchanda364]

Thank you for the help.

Answer 1

[Mensur Dlakic]

hmmscan scores a database of HMMs vs. a database of sequences. An HMM database is what needs to be processed by hmmpress.

Download:

wget http://dmk-brain.ecn.uiowa.edu/pVOGs/downloads/Ampullaviridae/AmpullaviridaevogHMMprofiles.tar.gz

Unpack:

tar -zxvof AmpullaviridaevogHMMprofiles.tar.gz

Concatenate HMMs into a database:

cat AmpullaviridaevogHMMprofiles/*.hmm > Ampullaviridae_all.hmm

Press:

hmmpress Ampullaviridae_all.hmm

Scan:

hmmscan Ampullaviridae_all.hmm your_sequences.fasta > search_output.txt

You may want to delete a directory with individual HMMs:

rm -rf AmpullaviridaevogHMMprofiles

[vishalchanda364]

Thank you so much sir. You're a gem, I was struggling with this since 3 days.

Thank you so much