Particle.news
Download on the App Store
3 ARTICLES
·
4d ago

AI-Driven Venom Screen Advances Antibiotic Peptide Candidates

University of Pennsylvania scientists are refining venom-derived peptide leads via medicinal chemistry with parallel safety and efficacy studies in animal models.

Image
Guan et al. demonstrate that venoms are a rich source of previously hidden antimicrobial scaffolds, and that integrating large-scale computational mining with experimental validation can accelerate the discovery of urgently needed antibiotics. Image credit: Guan et al., doi: 10.1038/s41467-025-60051-6.

Overview

  • The APEX deep-learning system scanned over 40 million venom-encrypted peptides in hours to flag 386 antibiotic candidates.
  • Researchers synthesized 58 AI-selected venom peptides and found 53 that killed drug-resistant Escherichia coli and Staphylococcus aureus without harming human red blood cells.
  • APEX also revealed more than 2,000 novel antibacterial motifs, expanding the structural diversity for drug development.
  • Preliminary in vivo tests have commenced to assess the efficacy and toxicity of top peptide leads in animal models.
  • Next steps focus on enhancing peptide stability and potency through medicinal-chemistry modifications ahead of advanced safety trials.