A Deep Learning-Driven Framework Integrating Organoid-Based Functional Validation Identifies Universal Neoantigens from Recurrent Glioma Mutations - PubMed
3 hours ago
- #glioblastoma
- #neoantigen
- #organoid
- Development of TCRscore, a neoantigen prediction algorithm integrating HLA binding and TCR recognition features.
- Validation using 21 patient-derived glioblastoma organoid models that retain parental tumor features.
- TCRscore outperformed six existing tools in predicting immunogenic neoepitopes.
- Organoid-T cell co-culture assays confirmed neoantigen-specific T cells induce targeted killing.
- Identification of recurrent PIK3R1G376R mutation as a potential shared neoantigen in glioblastoma.