Researchers announced on Apr. 6 the creation of the Extreme Environment Microbiome Catalog (EEMC), a new resource that uses artificial intelligence to identify potential antimicrobial peptides from some of Earth's harshest habitats.
The EEMC aims to address the growing challenge of antimicrobial resistance and the slowdown in antibiotic discovery by tapping into the largely unexplored genetic diversity found in extreme environments such as deep-sea vents, cryospheric regions, hypersaline lakes, geothermal sites, subsurface areas, and hyperarid systems. By doing so, scientists hope to discover new compounds that could lead to next-generation antibiotics.
According to a study published in Nature Communications, researchers reconstructed over 78,000 microbial genomes from thousands of metagenomes and isolates collected worldwide. They then used machine learning tools combined with large language models to screen for candidate antimicrobial peptides (cAMPs). The catalog enabled identification of thousands of these candidates; many showed activity against Gram-negative pathogens when tested in vitro. "Strikingly, over 86% of these species did not map to the comparison reference genome sets," researchers said. This suggests more than 20,000 potentially novel species have been identified.
The team also discovered nearly four billion unique genes within their samples—about one-fifth were unannotated in current databases—and identified more than 163,000 biosynthetic gene clusters with high novelty scores. Many genes were linked to stress adaptation and metabolic regulation. In experimental tests using synthesized peptides predicted by AI models, most inhibited bacterial growth without causing significant toxicity in mammalian cells.
Structural analysis revealed that many active peptides form α-helical shapes and disrupt bacterial membranes—a mechanism considered promising for fighting drug-resistant bacteria. One lead candidate peptide showed reduced likelihood for bacteria developing resistance over time.
The EEMC is positioned as a global reference platform for future exploration into microbial diversity and biosynthetic potential across extreme habitats. Researchers say further advances—such as improved sequencing technologies and expanded functional validation—could accelerate discovery efforts even more: "Looking ahead, integrating advanced sequencing technologies, artificial intelligence, and targeted cultivation strategies will be key... positioning the EEMC as a critical platform for future innovations in biotechnology and biomedicine."
