Fig. 2

IDH-wt immune clustering showed five distinct clusters associated with immune cell proportions and key genomic aberrations. a UMAP clustering of seven tumor subtypes of IDH-wt tumor cohort (N = 2072) based on immune cell proportion (Non-cancer part scaled from 0 to 1) showed subtype specific clustering. b Five optimum number of cluster obtained by using k-means clustering with Nbclust method. c Six major Immune cell type proportions shown by boxplots for each cluster indicated significant difference between immune cells distribution across all cluster. Y axis represents non-cancer cell proportion of particular immune cell scaled from 0 to 1. Each box plots depicted, boxes indicate interquartile range with central bar indicating median and whiskers indicating the range. Blue dot represents mean value of the proportion. T-test and Wilcoxon test (p value shown) were used to calculate statistical significance. d Sample proportions of seven tumor subtypes in each cluster of IDH-wt tumor cohort (N = 2072). Distribution of tumor subtypes in each cluster calculated as percentage of samples represent specific tumor subtype. e Cluster specific genomic aberrations represented as proportion of samples undergoes for genomic changes. Horizontal bars between cluster bars with asterisks represent chi- square test based comparison with significant p value < (0.0001, 0.001, 0.01, 0.05, 1; symbols = "****", "***", "**", "*", "ns”). f Monocyte proportion shown by boxplots for various sample groups with genomic aberrations. T-test and Wilcoxon test (p value shown) were used to calculate statistical significance