These data were used for prediction of neoantigens and minor histocompatibility mismatch antigens, which were subsequently used for design of a machine-learning algorithm for tumor-specific antigen (TSA) immunogenicity prediction.  TSA vaccines are a growing area of study for cancer immunotherapy, but identification of clinically relevant targets remains a challenge. The study associated with these data provides the first description of a computational method for direct prediction of TSA immunogenicity trained entirely from validated TSA immunogenicity scores. This tool has allowed us to 1) predict for clinically efficacious TSA targets, 2) identify genomic correlates of TSA immunogenicity, and 3) demonstrate evidence of alternative out-of-frame TSAs which can promote anti-tumor immunity. SOURCE: Benjamin,Garrett,Vincent (benjamin_vincent@med.unc.edu) - Vincent Lab University of North Carolina at Chapel Hill

Pluto Bioinformatics

GSE136619: Whole exome and transcriptome sequencing of murine tumor models