AntiSMASH genome mining for novel hybrid polyketide-nonribosomal peptide-specialized lipids output

This repository contains results from an antiSMASH (version 7.1.0) genome mining approach on a dataset comprising 62,706 NRPS-containing bacterial genomic regions in Genbank format, obtained from the antiSMASH database (version 4). The dataset was analyzed via a custom-designed search rule, that we termed ‘zeamine-like’, with the goal of finding novel trihybrid polyketide synthase (PKS)-nonribosomal peptide synthetase (NRPS)-polyunsaturated fatty acid (PUFA) synthase-like clusters. This rule integrated pre-existing profile Hidden Markov Models (pHMMs) targeting PKS, NRPS and PUFA synthase-like biosynthetic machinery (cds(Condensation and (AMP-binding or A-OX)) and cds(PKS_AT and (PKS_KS or ene_KS or mod_KS or hyb_KS or itr_KS or tra_KS)) and (hglE or hglD or PUFA_KS)). A ‘relaxed’ level of strictness was applied, allowing for the detection of incomplete clusters lacking one or more functional components. The maximum allowed distance between core genes was set at 20 kbp, and an additional 20 kbp was included beyond the core genes to define the protocluster boundaries. This repository contains the combined results output from antiSMASH.