.png){kind=icon}
.png){kind=icon}
Return
AI-driven design of antimicrobial peptide for combating resistance and infectious diseases
.png){kind=icon}
PRE
AI
DOI：10.1016/j.prp.2026.156445.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The growing threat of antimicrobial resistance, coupled with the challenges of developing new antibiotics, demands innovative therapeutic solutions. Antimicrobial peptides (AMPs) present a promising alternative, yet their clinical application is limited by toxicity, instability, low permeability, and high production costs. To overcome these barriers, we employed artificial intelligence (AI) and machine learning (ML) to design a fifteen-amino acid peptide, LCN-15 (RWWRRKKLKAPIWVR), with a molecular weight of 2079 Da. This short cationic peptide, rich in arginine and tryptophan residues, exhibits strong membrane interaction and antimicrobial potential. Using AIguided de novo design, we rapidly analyzed structural features, predicted biological activities, and optimized the sequence for enhanced safety and efficacy. ML-based predictive assays indicated broad functional potential, encompassing anti-microbial, anti-biofilm, anti-cancer, and anti-oxidant properties. Subsequent analysis indicated favourable safety attributes, including low toxicity, minimal hemolytic potential, and good blood brain barrier permeability. In addition, predictive models suggested potential immunomodulatory activity, indicating that LCN-15 may enhance host defense mechanisms alongside its direct anti-microbial effects. Further, to correlate our predictive modelling of LCN-15, we used melittin (GIGAVLKVLTTGLPALISWIKRKRQQ), a 26 amino-acid cationic linear AMP which is very well studied. This comprehensive in silico predictive analysis, performed prior to peptide synthesis, ensured that only the most promising designs were advanced for consideration, thereby streamlining the workflow, minimizing experimental steps, and reducing overall costs. Furthermore, the consistency between the predicted activities of melittin and its well-established in vitro properties further supports the reliability of the computational predictions.These findings position LCN-15 as a multifunctional therapeutic candidate with potential applications in managing infections, modulating immune responses, and addressing the urgent global challenge of antimicrobial resistance.
Keywords:
Anti-Microbial Peptides
AMR
Designing
Engineering
Infectious Diseases
Journal
P
IF:
3.2
Papers:
463
Citations:
9.8K