(hmmsearch) find targets with multiple query profiles with hmmsearch

Question

[usr2]

hmmsearch manual specifies:

"hmmsearch is used to search one or more profiles against a sequence database"

However I could not find specification about how to specify multiple queries. I tried hmmsearch [options] <hmmfile1> <hmmfile> <seqdb> without succes.

Does someone know if its possible to search for proteins with two profiles?

thanks & cheers,

Answer 1

[Mensur Dlakic]

You will have to concatenate multiple HMMs into a single file. They have a double slash // to signify the end of profile, and the program knows that whatever comes next will be a separate profile. After that, you do a simple command:

hmmsearch [options] <concatenated_hmmfile> <seqdb>

[usr2]

Many thanks! :)

[nramakrishna837]

how can we download seqdb file

[GenoMax]

You can use any appropriate multi-fasta sequence file.

[nramakrishna837]

I have one protein fa file. Now how can I convert that file into seqdb?

For hmmsearch, I wrote one command line once please check and let me know what's wrong in it hmmsearch --tblout <profile_PF03936.hmm> GCF_907164915.1_protein.faa > search_PF03936

[GenoMax]

Now how can I convert that file into seqdb?

No conversion is needed.

Here is a basic example courtesy of GPT. You need to have these minimum options.

hmmsearch -o output.txt --tblout results.tbl HMM_model.hmm sequence_database.fasta

hmmsearch: The command to run HMMER’s hmmsearch tool.
-o output.txt: This specifies the output file where the standard results will be written.
--tblout results.tbl: This option specifies a tabular output file for the results, which is easier to parse and contains the summary of hits.
HMM_model.hmm: The input file containing the HMM profile model you want to search with.
sequence_database.fasta: The FASTA file containing the sequence(s) to be searched against the HMM profile.

[Mensur Dlakic]

hmmsearch --tblout <profile_PF03936.hmm> GCF_907164915.1_protein.faa > search_PF03936

You should not be literal with the example command I gave. Left and right single angle characters in the example above signify that the name inside is generic, and should be replaced with your own appropriate file name. For your example it most likely should be:

hmmsearch --tblout table_out.txt profile_PF03936.hmm GCF_907164915.1_protein.faa > search_PF03936

This will use a HMM file profile_PF03936.hmm, a sequence file GCF_907164915.1_protein.faa and save tabular output in table_out.txt and the screen output in search_PF03936. That means you didn't provide a name for the table output and HMM name should not be in single angle brackets.