Welcome to DLpTCR
Most of the antigen peptides predicted in silico fail to elicit immune responses in vivo. Consequently, it is necessary to develop novel computational methods for accurately predicting immunogenic peptide recognized by T cell receptor (TCR), thereby helping vaccine development and cancer immunotherapies. Here, we described DLpTCR, a multimodal ensemble deep learning framework for predicting the likelihood of interaction between a single or paired chain of TCR and peptide presented by major histocompatibility complex (MHC) molecules. The DLpTCR model exhibits high predictive performance, demonstrating the models' ability to learn general interaction rules and generalize to antigen peptide recognition by TCR.
