Figure 3
From: RNA Transcription and Splicing Errors as a Source of Cancer Frameshift Neoantigens for Vaccines
Detecting of antibody response against FS in cancer patients. (A) Design of human FS array with microsatellite FS peptides from all coding MS regions and predicted mis-splicing FS peptides from every exon of human genes. (B) Common reactivity and cancer-type reactivity against FS peptides were represented by ~7000 selected FS peptides. LC: lung cancer; BC: breast cancer; GBM: glioblastoma; GC: gastric cancer; and PC: pancreatic cancer (n = 17/each cancer type) and a set of non-cancer samples (n = 64), as control. (C) p-value and fold change volcano plot analysis of 5 cancer’s IgG reactivity on the 400 K FS array compared to normal. The red line represents the significant p-value cut-off = 1/392328 (the number of the array peptides). (D) Positive rate of all 400 K FS peptides in each cancer type, overall cancer and normal group (calculated by counting samples with higher reactivity than Average(Normal) + 6*SD (Normal)), error bar represents Mean ± SEM. (E) Distribution of personal anti-FS response and shared anti-FS response in all 5 cancer types. (F) Top 20 FS peptides for each GBM sample were selected for personal vaccines. (G) Components of cancer-type specific FS vaccines, top 100 FS peptides for each cancer type were selected with highest positive rate in corresponding cancer type. Red: positive sample, blue: negative sample. (H) Components of a general FS vaccine, top 100 FS peptides were selected with highest positive rate in cancer group. Red: positive sample, blue: negative sample. (I) Heatmap of the positive rate distribution of the FS peptides in Stage I and late stages pancreatic cancer.
