Fig. 4

Neoantigens derived from each isoform or mutation type in cell lines. a The number of neoantigen candidates in each cell line. b The distribution of the number of novel peptides derived from tumor-specific regions of each isoform or mutation type. c The distribution of the number of neoantigen candidates from b. d The distribution of the maximum NetMHC score for each isoform or mutation type. ***P < 0.001 and **P < 0.01 (Kruskal–Wallis test and Dunn–Bonferroni’s post hoc test). e A correlation between the gene expression levels calculated by short-read RNA sequencing (TPM) and the number of peptides detected by proteome analysis for each gene in RERF-LC-Ad1 (r = 0.52). Both values + 1 were log2-transformed. Red points represent the KRT7 gene with the isoform whose peptides were detected by proteome, the green area shows 0-TPM genes, and the blue area shows 0-peptide genes. f A comparison of genes detected by RNA sequencing (TPM ≥ 1) and by proteome (≥ 1 peptide) in RERF-LC-Ad1. Genes (4021/4064, 99%) detected by proteome were covered by RNA sequencing. g, h The full-length structure of the splicing isoform of KRT7 (g) and the magnified inset of the alternative 5′ splice site region (h). The detected peptide of “DRLEEGCLK” was derived from exon 4 in the aberrant isoform