enFoldX reaches AUC 0.82 on human VDJdb and transfers to mouse at 0.76
quaidmorris · x · 2026-07-22
enFoldX reports strong cross-species transfer on VDJdb
The post adds that on VDJdb ten-fold cross-validation, enFoldX reaches AUC 0.82 on human TCR-split evaluation and 0.98 on mouse.
A notable result is transfer learning: a model trained only on human data still classifies mouse TCR:pMHC pairs at AUC 0.76, which the author describes as a first for this task.
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