Artificial Intelligence for Therapeutic Peptide Discovery and Development
- 1Institut de Biotecnologia i de Biomedicina, Departament de Bioquímica i Biologia Molecular, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, 08193 Barcelona, Spain
Abstract
Therapeutic peptides combine biological specificity with chemical versatility, and artificial intelligence (AI) now contributes to their identification, design, and optimisation. Peptide discovery, however, cannot be reduced to generating amino-acid sequences: therapeutic performance also depends on chemistry, conformation, assay context, stability, delivery, safety, manufacturability, and in vivo exposure. This critical review examines AI use across the major peptide classes: receptor agonists, protein–protein interaction inhibitors, oncology and tumour-targeting peptides, antimicrobial and phenotype-driven peptides, delivery peptides, vaccine epitopes, self-assembling materials, and chemically constrained macrocycles. It compares sequence-based, chemical, graph, structure-aware, and multimodal representations across predictive, generative, and experimentally integrated workflows. Coverage spans recent general protein foundation, prediction, and design systems (ProtGPT2, ESM-2/ESM3, ProGen/ProGen2, ProteinMPNN, RFdiffusion, and AlphaFold 3) and peptide-focused methods (RFpeptides, PepFlow, and PepGLAD). Particular attention is given to benchmark integrity, the limits of structure prediction for short and non-canonical molecules, and the gap between retrospective and prospective evidence. To make claims comparable, the review adapts established reporting, validation, and readiness concepts into a peptide-specific, two-dimensional Peptide–AI Evidence Grid that separates methodological rigour (M0–M3) from translational depth (T0–T3), applies it to representative studies, and provides a checklist extending DOME, TRIPOD + AI, and FAIR guidance with the chemistry- and assay-specific items peptide work requires.
Related
- Artificial Intelligence in Mass Spectrometry-Based Metabolomics and Lipidomics for Drug Discovery and Development
- Bone morphogenetic protein 4 meets artificial intelligence: a comprehensive review of AI-driven advances in drug discovery, structural modeling, and clinical translation
- Artificial Intelligence in Virus Detection: From Metagenomic Discovery to Clinical Decision Support and Population Surveillance.
- Discovery of an NRAS isoform and activation-state selective macrocyclic peptide
- A genome-wide computational approach for therapeutic targets identification and synthetic peptide assembly enable rational vaccine design against Arcanobacterium haemolyticum
BibTeX
@article{ventura2026artificial,
title = {{Artificial Intelligence for Therapeutic Peptide Discovery and Development}},
author = {Salvador Ventura},
journal = {Biomolecules},
year = {2026},
month = oct,
doi = {10.3390/biom16101445},
url = {https://www.mdpi.com/2218-273X/16/10/1445},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Artificial Intelligence for Therapeutic Peptide Discovery and Development
AU - Salvador Ventura
PY - 2026
DA - 2026/10/03
JO - Biomolecules
DO - 10.3390/biom16101445
UR - https://www.mdpi.com/2218-273X/16/10/1445
AB - Therapeutic peptides combine biological specificity with chemical versatility, and artificial intelligence (AI) now contributes to their identification, design, and optimisation. Peptide discovery, however, cannot be reduced to generating amino-acid sequences: therapeutic performance also depends on chemistry, conformation, assay context, stability, delivery, safety, manufacturability, and in vivo exposure. This critical review examines AI use across the major peptide classes: receptor agonists, protein–protein interaction inhibitors, oncology and tumour-targeting peptides, antimicrobial and phenotype-driven peptides, delivery peptides, vaccine epitopes, self-assembling materials, and chemically constrained macrocycles. It compares sequence-based, chemical, graph, structure-aware, and multimodal representations across predictive, generative, and experimentally integrated workflows. Coverage spans recent general protein foundation, prediction, and design systems (ProtGPT2, ESM-2/ESM3, ProGen/ProGen2, ProteinMPNN, RFdiffusion, and AlphaFold 3) and peptide-focused methods (RFpeptides, PepFlow, and PepGLAD). Particular attention is given to benchmark integrity, the limits of structure prediction for short and non-canonical molecules, and the gap between retrospective and prospective evidence. To make claims comparable, the review adapts established reporting, validation, and readiness concepts into a peptide-specific, two-dimensional Peptide–AI Evidence Grid that separates methodological rigour (M0–M3) from translational depth (T0–T3), applies it to representative studies, and provides a checklist extending DOME, TRIPOD + AI, and FAIR guidance with the chemistry- and assay-specific items peptide work requires.
SN - 2218-273X
ER -