Align protein sequence to Pfam file using HMMER3.0 (hmmalign)

0

Entering edit mode

6.2 years ago

Xiaokang ▴ 70

I'm using HMMER 3.0 (hmmalign to be specific) to align the whole proteome of Human to Pfam file. I got the proteome from https://www.ensembl.org/info/data/ftp/index.html, got the Pfam file from: ftp://ftp.ebi.ac.uk/pub/databases/Pfam/releases/Pfam31.0/, and downloaded the file Pfam-A.hmm.gz

With the command

hmmalign -o human.out Pfam-A.hmm Homo_sapiens.GRCh38.pep.all.fa

I got the error:

Error: HMM file Pfam-A.hmm does not contain just one HMM

HMMER protein Pfam • 2.8k views

ADD COMMENT • link updated 6.0 years ago by Biostar 20 • written 6.2 years ago by Xiaokang ▴ 70

0

Entering edit mode

You might want to try hmmsearch or hmmscan (for the latter, you need to hmmpress your Pfam-A first).

ADD REPLY • link 6.2 years ago by cschu181 ★ 2.8k

0

Entering edit mode

Thank you. But why hmmscan, instead of hmmsearch? Since I found an article comparing them here. It seems that they have the same output, but hmmsearch is faster?

ADD REPLY • link 6.2 years ago by Xiaokang ▴ 70

0

Entering edit mode

I have always asked this myself. I know, the output is not identical, at least not when you use the tblout/domtblout options. hmmscan is domain/hmm-oriented, whereas hmmsearch is sequence oriented. In the end, the speed issue as mentioned in your link (nice read by the way!) would convince me to use hmmsearch and then reformat the output.