Researchers unveil UpTCR model for TCR-antigen binding prediction
A team of researchers has developed UpTCR, a foundation model designed to predict T-cell receptor (TCR) binding to antigenic peptides presented by HLA molecules. Published in Nature Communications, the model addresses data scarcity by progressively transferring knowledge from simpler dimeric and trimeric interactions to complex tetrameric structures. It employs soft contrastive learning to…
Key points
- UpTCR predicts TCR-antigen-HLA binding using progressive knowledge transfer from incomplete data
- Model identifies 8 immunogenic melanoma peptides and 1 immune escape variant in validation
- Researchers include Tencent AI for Life Sciences Lab employees; funded by Chinese grants
The model outperforms existing methods in predicting binding specificity and affinity, particularly for neoantigens. In prospective validation using melanoma antigen variants, UpTCR identified eight immunogenic peptides that triggered T-cell responses and one variant linked to immune escape. The authors, including employees of Tencent AI for Life Sciences Lab, report the tool generalizes to breast cancer cohorts with limited data and reveals residue-level interaction maps. Funding came from multiple Chinese national and regional grants.
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Microsoft Copilot AI predicts bitcoin could reach $180,000 by early 2027 · 2 src
- MIT researchers develop tool to estimate suicide risk from Text messages · 1 src
- Anthropic's Claude discovers new enzyme system ART resembling crispr · 17 src
- Study finds AI experts underestimated progress on benchmarks and adoption · 1 src
- Sakana AI hires Schmidhuber as chief scientific advisor · 1 src
Comments
via GitHub Discussions