enFoldX classifies TCR–pMHC pairs from AF3 ensembles instead of one structure
quaidmorris · x · 2026-07-22
enFoldX classifies TCR:pMHC pairs from AF3 ensembles
The diagram shows the pipeline:
- Run AF3 on TCRα, TCRβ, MHC, and peptide inputs.
- Generate a structural ensemble rather than a single prediction.
- Extract interface and confidence features from the ensemble.
- Aggregate them with mean and standard deviation.
- Feed the features into a classifier to separate cognate from non-cognate pairs.
The thread argues that useful signal lives in how often AF3 hallucinates different structures, not just in the top-ranked model.
Related event: enFoldX Turns AlphaFold3 Noise into TCR Recognition Signals(11 posts)→
More from Research
- EdgeRock-4B holds KV cache at 72 MiB across 256K context, 99.1% smaller than Qwen3.5-4B — alifcoder · 2026-10-27
- TRL adds Score Centering, a simple fix for RLHF training instability — lvwerra · 2026-10-27
- Latent-Foresight: world models that learn to predict their own future — abursuc · 2026-10-27
- UK AISI papers: no good safeguards yet against misuse of biological AI tools — davidmanheim · 2026-10-27
- CMU's XYBench: LLMs answer queries literally and reinforce user misconceptions — valentina__py · 2026-10-27
- arXiv advice from T. Dietterich: wrong primary category will almost surely get your paper on hold — tdietterich · 2026-10-27