Design of Cyclic Peptides Targeting Protein–Protein Interactions Using AlphaFold
Problem
Predicting structures of protein–cyclic peptide complexes and designing cyclic peptides that bind target proteins was computationally infeasible, blocking cyclic peptides as a modality for protein-protein interaction drug targets.
Why it matters
Protein-protein interactions are numerous major therapeutic targets that conventional small molecules struggle to drug.
What the AI did
The authors added a cyclic peptide complex offset to AlphaFold2, enabling structural prediction of protein–cyclic peptide complexes, and applied it to AfDesign's binder hallucination protocol for de novo cyclic peptide design.
Results
not stated in source (only qualitative claim of 'high accuracy' prediction)
Impact
Authors report high-accuracy complex prediction enabling de novo cyclic peptide design; no numbers given.
Record
| Venue | International Journal of Molecular Sciences |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Takatsugu Kosugi, Masahito Ohue |
| Citations | 51 (OpenAlex, at import time) |
| References | 69 |
| Model or method | AlphaFold2 (cyclic peptide complex offset); AfDesign |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/ijms241713257 |
| Open access | gold |
| Keywords | AlphaFold, cyclic peptides, protein-protein interactions, de novo design, AfDesign |
Importance rating
significant — thousands to millions affected
Abstract quantifies >930,000 identified PPIs as major therapeutic targets poorly addressed by small molecules; enables a genuinely new design capability.