TCRAD: An End-to-End Framework for Antigen-Targeted T Cell Receptor Design
Published in bioRxiv(Under review in Briefings in Bioinformatics), 2026
TCRAD is an end-to-end deep learning framework for de novo design of antigen-specific TCR CDR3β sequences. The framework integrates sequence generation, binding filtration, and structure prediction modules. In experimental validation using the 1G4/NY-ESO-1 system, 17.2% (5 of 29) of the designed candidate TCRs elicited antigen-induced activation in NFAT-reporter J76 cells.
Recommended citation: Chenao Li, Yaochi Guo, Xin Guan, Hui Chen, Yong Zhang, Pengyuan Yang, Jizhong Lou. (2026). "TCRAD: An End-to-End Framework for Antigen-Targeted T Cell Receptor Design." bioRxiv.