AI Searches for New Antibiotics in Animal Venoms

A Hidden Treasure

According to the authors of a study published in Nature Communications, animal venoms represent a vast and largely untapped reservoir of bioactive molecules with antimicrobial potential. They are rich in bioactive peptides and proteins that exhibit a range of effects, including antibacterial activity. Many venom peptides disrupt cell membranes, making it difficult for bacteria to develop resistance.

In addition, these peptides often show broad-spectrum activity against both gram-positive and gram-negative bacteria, making them a promising option in the fight against multidrug-resistant pathogens. According to one of the study’s authors, César de la Fuente, venoms are “evolutionary masterpieces” whose antimicrobial potential remains largely unexplored.

The Golden Mean

Previous studies have suggested that peptide-based antimicrobial agents represent an attractive alternative to both traditional antibiotics and protein-based therapeutics. While small-molecule antibiotics are highly effective, they carry a high risk of rapid resistance development.

Proteins, on the other hand, can achieve greater specificity but are limited by issues such as stability and bioavailability. Antimicrobial peptides—including those derived from venoms—could represent a golden middle ground, combining potent antimicrobial activity with structural flexibility and modifiability.

Using the APEX System

Researchers from the University of Pennsylvania employed a deep learning system called APEX. According to the study’s authors, APEX can scan vast chemical spaces within hours and identify peptides with exceptional potential to combat some of the world’s most resistant pathogens. This reduces dependence on resource-intensive biochemical testing.

Using this system, scientists examined over 16,000 venom proteins from snakes, scorpions, spiders, and other organisms, generating more than 40 million potential venom-derived peptides. These are small proteins evolved by animals for attack or defense. Within a very short time, the algorithm identified 386 compounds with molecular features characteristic of next-generation antibiotics.

From this narrower AI-generated selection, the research team synthesized 58 venom peptides for laboratory testing. Of these, 53 proved effective against bacteria resistant to currently available drugs, including Escherichia coli and Staphylococcus aureus. Moreover, the doses used were harmless to human red blood cells.

A Solution for the Future

By combining computational analysis with rapid laboratory validation, scientists conducted one of the most comprehensive studies to date on venom-derived antibiotics. According to the authors, the results support the hypothesis that venom-derived peptides not only retain their antimicrobial function when isolated from their parent toxins but may also serve as templates for future peptide-based drugs.

(hno)

Sources:

1. Guan C., Torres M. D. T., Li S. et al. Computational exploration of global venoms for antimicrobial discovery with Venomics artificial intelligence. Nat Commun. 2025 Jul 12;16 (1): 6446, doi: 10.1038/s41467-025-60051-6.

2. Murray C. J. L., Ikuta K. S., Sharara F. et al. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet 2022 Feb 12; 399 (10325): 629–655, doi: 10.1016/S0140-6736(21)02724-0.

3. Magana M., Pushpanathan M., Santos A. L. et al. The value of antimicrobial peptides in the age of resistance. Lancet Infect Dis. 2020 Sep; 20 (9): e216–e230, doi: 10.1016/S1473-3099(20)30327-3.

4. de la Fuente-Nunez C., Torres M. D., Mojica, F. J. et al. Next-generation precision antimicrobials: towards personalized treatment of infectious diseases. Curr Opin Microbiol 2017 Jun 37 : 95–102, doi: 10.1016/j.mib.2017.05.014.

5. Ciulla M. G., Civera M., Sattin S. et al. Nature-inspired and medicinally relevant short peptides. Explor. Drug Sci, doi: 10.37349/eds.2023.00011.

6. Wang L., Wang N., Zhang W. et al. Therapeutic peptides: current applications and future directions. Signal Transduct Target Ther 2022 Feb 14;7 (1): 48, doi: 10.1038/s41392-022-00904-4.

7. Rossino G., Marchese E., Galli G. et al. Peptides as Therapeutic Agents: Challenges and Opportunities in the Green Transition Era. Molecules 2023 Oct 19;28 (20): 7165, doi: 10.3390/molecules28207165.