Abstract Juvenile presentation of HGG has been shown to frequently involve somatic mutations of the H3-3A gene, which encodes a histone H3 variant. To address the paucity of identified TSAs in DHG, we utilized a novel computational platform, Isoform peptides from RNA splicing for Immunotherapy target Screening (IRIS). The IRIS pipeline is designed to leverage large-scale cancer and normal transcriptomic data to identify peptide sequences that result from alternative splicing events that are highly enriched in cancer cells, and are predicted to bind the HLA class I alleles carried by the individual patient. To investigate whether peptide-pulsed dendritic cells (DC) together with checkpoint blockade might be a potential treatment modality for diffuse hemispheric glioma, H3 G34-mutant, we have developed a novel mouse model, and utilized a novel bioinformatics pipeline to identify tumor-associated peptide antigens. We used the RCAS/tv-A system to target the expression of H3/G34R and PDGF-B and to knock out p53 in neural progenitors in transgenic C57BL/6 neonatal mice. Three independent cell lines were obtained that expressed transcripts associated with oligodendrocyte and interneuron lineages, and formed lethal tumors after intracranial implantation. Implementing IRIS, we identified nonapeptides generated from alternative splicing of mRNAs that were highly enriched in the tumor transcriptomes of two H3/G34R patients, and predicted to bind the HLA-A*2:01 class I allele carried by these patients. Three of the candidate peptides were conserved in the mouse and also predicted to bind the H2-Kb MHC I allele carried by C57/BL6 mice. Using our syngeneic mouse model, we showed that DCs pulsed with these peptides together with anti-PD1 mAb provided a significant survival benefit compared with checkpoint inhibitor and DCs pulsed with peptides that were not predicted to bind the H2-Kb allele.