A Deep Learning-Driven Framework Integrating Organoid-Based Functional Validation Identifies Universal Neoantigens from Recurrent Glioma Mutations

A novel neoantigen prediction algorithm, TCRscore, based on publicly available datasets by integrating human leucocyte antigen binding and T-cell receptor (TCR) recognition features. TCRscore was developed based on publicly available datasets, and incorporates conventional features, such as MHC binding affinity and gene expression, as well as TCR recognition potential, aiming to better capture the true immunogenicity of candidate peptides.