PRISM introduces uncertainty quantification to TCR specificity prediction, addressing viral bias in public datasets through synthetic data augmentation.bioRxiv
This methodological advance offers a more robust framework for identifying rare epitopes where traditional models often fail.bioRxiv
This item is a correction notice regarding the repurposing of public sarcoma multi-omics data for neoantigen discovery. It signals an update to prior work but provides no new clinical or methodological details in this brief.
PRISM is a metric-learning framework that embeds TCR beta sequences into a peptide-organized latent space to predict pMHC ligands. By using structure-guided synthetic receptors to counter viral bias in training data, it abstains on out-of-distribution inputs via uncertainty modeling. Benchmarks show PRISM matches or exceeds existing models, particularly for rare epitopes.