Fig. 6

Discrimination of potential immunotherapeutic targets in DLBCL. a–c DLBCL MAPs, identified through a TSA-discovery proteogenomic approach, were searched with BamQuery in GTEx tissues (n = 12–50 / tissue), mTECs (n = 11), sorted blood B cells (n = 14), our DLBCL specimens (n = 3), and TCGA DLBCL (n = 48) bam files in unstranded mode with genome version GRCh38.104 and dbSNP version 155. a Heatmap of average RNA expression of 67 TSA candidates in indicated tissues. Boxes in which a peptide has an rphm > 8.55 are highlighted in black. b Heatmap of average RNA expression of the highest shared and expressed TSA candidates (11) in cancer samples DLBCL from TCGA (n = 48). Boxes in which MAPs expression (rphm) is > 8.55 are highlighted in black. c Percentage of the most likely biotype attributed by BamQuery for TSA candidates (n = 67). d Repitope immunogenic scores calculated for negative control thymic MAPs (n = 158), highly expressed DLBCL TSAs (n = 18, 25% of TSAs most upregulated by DLBCL TCGA versus normal blood in GTEx and sorted B cells), and positive control HIV MAPs (n = 450). Mann–Whitney U test was used for comparisons (*p < 0.05, ****p < 0.0001). e Pearson’s correlation in TCGA DLBCL patients (n = 48) between the count of highly expressed (HE) TSAs expressed by each patient and the expression of cytotoxic T cell markers (CD8A + CD8B, in counts per million (cpm)). The red line is a linear regression. f Network analysis of GO term enrichment among genes overexpressed by patients expressing an above-median number of HE-TSAs. Line color reflects the similarity coefficient between connected nodes. Node color reflects the false discovery rate (FDR) of the enrichment. Node size is proportional to gene set size