Custom Hidden Markov Models (HMM) for methylmercury detoxification genes merA and merB.

We built custom Hidden Markov Models for mercury reduction (merA) and methyl-mercury demethylation (merB) genes. Briefly, we queried the KEGG ortholog identifiers, K00520 and K00221, in UniProtKB to retrieve merA and merB sequences, obtaining a total of 404 and 55 non-redundant amino acid sequences respectively. The function of the sequences was verified by checking the presence of the amino acids considered to be essential for their activity (377 for merA and 51 for merB). Verified sequences were aligned in MAFFT (default parameters) and we used hmmbuild in HMMER v3.3 to create the 2 HMM profiles.