De Novo Multi-Mechanism Antimicrobial Peptide Design via Multimodal Deep Learning - PubMed
3 days ago
- #deep learning
- #drug resistance
- #antimicrobial peptides
- AI-driven discovery of antimicrobial peptides (AMPs) using 3D structural characteristics, species-specific activities, and mechanisms.
- QLAPD database created with sequences, structures, and antimicrobial properties of 12,914 AMPs.
- M3-CAD pipeline proposed for de novo design of multi-mechanism AMPs to combat multidrug-resistant organisms (MDROs).
- Pipeline integrates generation, regression, and classification modules with a 3D voxel coloring method.
- QLX-3DV-1 and QLX-3DV-2 identified through M3-CAD show multiple antimicrobial mechanisms, activity against MDROs, and low toxicity.
- In vivo experiments validate antimicrobial effects with limited toxicity.
- Integration of 3D features enhances AI-driven AMP discovery.