Abstract

The emergence of multifaceted diseases and antibiotic-resistant organisms has demonstrated the necessity to develop new therapeutic candidates, especially those of natural ecosystems. The current study integrates molecular ecology and the recent biotechnological approaches, such as metagenomics, metabolomics, and analyses of environmental DNA, to research on the biosynthesis of bioactive metabolites within the ecologically-influenced environment. It aims to create an ecology-aware model of drug discovery by discovering bioactive compounds that are influenced by evolutionary pressures across habitats. We integrated 18,462 species occurrence records, 1,204 documented biotic interactions, and 312 environmental (BGC) enrichment. From genomic and metabolomic datasets, 4,892 BGCs were identified, with NRPS (42%) and Type I PKS (31%) being the most abundant. The functional annotation of these clusters indicated that these clusters were 63 percent involved in antimicrobial and cytotoxic compounds biosynthesis, and 57 percent of them were environmentally induced. The metabolomic data suggested the occurrence of 1,362 different metabolites; the metabolites of the adaptive stress-adapted taxa were enriched with phenolics, siderophores, and lipid-derived compounds. Bioactivity profiles indicated 48% and 27% of metabolites had antimicrobial and anticancer activities, respectively. The machine-learned models (accuracy = 0.87, F1 score = 0.83) were capable of predicting bioactivity, and molecular docking provided high binding affinities to infection-related, inflammation-related, and cancer-related targets. The study is based on the potential of ecology in drug studies, in which the hypothesis is that there is a need to put more emphasis on stressful environments to find new therapeutic agents.