hmmsearch

[cnatalie]

rule run_hmmer in annotate-snake should use the transdecoder .pep output as the fasta_file input instead of the mmseq-clustered nucleotide assembly, because the 6 frame translation is way too slow and does not predict proteins well(!!)