Structural ensembles beat single predictions in TCR:pMHC generalization study
quaidmorris · x · 2026-07-22
The takeaway from this post is that structural ensembles, not single predicted structures, contain the biophysical signal needed for robust generalization in TCR:pMHC prediction.
- The same idea is claimed to extend beyond AF3 to Boltz, RoseTTAFold, OpenFold, and related models.
- A follow-up note says SHAP highlights key drivers such as iPAEs, iPTMs, and pLDDT.
- The peptide:MHC interface also matters, suggesting both successful presentation and TCR recognition contribute.
- Ensemble means and standard deviations each add independent signal.
Related event: enFoldX Turns AlphaFold3 Noise into TCR Recognition Signals(11 posts)→
More from Research
- rauchg reproduces OpenAI's covert CPU-channel finding: chat across Firecracker microVMs with no network — ccerrato147 · 2026-10-27
- Microsoft Research: ForkPilot cuts long-horizon agent compute by up to 59% — EliasEskin · 2026-10-27
- Microsoft's TeleTune evolves agent skills from offline user telemetry — EliasEskin · 2026-10-27
- FlexGlue: sparse-to-dense matching that beats dense accuracy at a fraction of the cost — zhenjun_zhao · 2026-10-27
- Mundo turns a few room photos into an explorable, simulatable 3D world — at 1.5 FPS — mark_k · 2026-10-27
- Mathematics as an experimental science: Marc Lelarge to talk LLM-driven math experiments at ENS — marc_lelarge · 2026-10-27