Deep generative modeling for protein design
importance 7/10
ai assisted
Protein Structure and Dynamics protein science stage: research
Problem
Designing novel proteins with desired structures, stability and function is slow and difficult; generative models are being applied to accelerate it.
Why it matters
Novel proteins could enable new medicines, enzymes and materials.
What the AI did
Review of five classes of deep generative models that learn protein representations and can rapidly propose millions of novel candidate proteins, guided by discriminative oracles.
Results
not stated in source
Impact
Review synthesizes progress in AI-driven protein design; no new result demonstrated.
Record
| Venue | Current Opinion in Structural Biology |
|---|---|
| Published | 2021 |
| Work type | review |
| Authors | Alexey V. Strokach, Philip M. Kim |
| Citations | 115 (OpenAlex, at import time) |
| References | 102 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1016/j.sbi.2021.11.008 |
| Open access | hybrid |
| Keywords | protein design, generative models, deep learning, protein sequences, review |
Importance rating
7
significant — thousands to millions affected
Protein design underpins potential new medicines, enzymes, and materials; generative models can propose millions of candidates, a genuinely new capability for a foundational field.