INFO on how to scan a FASTA file with Gene3D HMM models resolve Domain Architectures. (The methods have been tested on hmmer v3.1b2) #First of all extract the HMM models TAR archive tar -xvf hmms.tar.gz #run hmmsearch on sequence file, change parameters as appropriate hmmsearch -Z 10000000 --domE 0.001 --incdomE 0.001 -o seqs.hmmsearch hmms/main.hmm seqs.fa #resolve any conflicting (overlapping) domain assignments, versions for different platforms (mac/linux) are available online along with documentation: # http://cath-tools.readthedocs.io/en/latest/tools/cath-resolve-hits/ # we recommend using a worst-permissible-bitscore of 25 and applying --min-dc-hmm-coverage=80 (only available for processing hmmsearch output) ./cath-resolve-hits --min-dc-hmm-coverage=80 --worst-permissible-bitscore 25 --output-hmmer-aln --input-format hmmsearch_out seqs.hmmsearch > seqs.crh #assign CATH superfamilies to the output file generated by cath-resolve-hits (in the future it is likely that this functionality will move into the cath-resolve-hits tool) python ./assign_cath_superfamilies.py seqs.crh