Fig. 1. Mutation density and timing are associated with age at diagnosis.

Summary of associations between age and (A) SNV density and (D) percent genome altered (PGA) in TCGA, PCAWG and AACR GENIE tumours. The dot size and colour show the Spearman correlation, and background shading indicate adjusted multivariate p value. Only tumour-types with at least univariately significant associations are shown. Associations between (B) SNV density and (E) PGA with age in prostate cancer across the three datasets (n_TCGA = 492, n_PCAWG = 199, n_AACR-GENE = 582 biologically independent samples). Univariate Spearman correlation, FDR-adjusted correlation p value and FDR-adjusted multivariable linear regression p values shown. C Correlations between age and proportion of SNVs occurring in the truncal clone in four PCAWG tumour contexts (Spearman correlation and linear regression FDR-adjusted p values). F In TCGA lung adenocarcinoma (n = 497 biologically independent samples), age and PGA are associated with smoking history (Kruskal–Wallis test), and (G) the negative association between PGA and age remains significant in current smokers and recently reformed smokers (≤15 years; Spearman correlation p values). From top to bottom: never smokers (yellow), current reformed smokers >15 years (green), current reformed smokers ≤15 years (light blue), current smokers (dark blue). Tukey boxplots are shown with the box indicating quartiles and the whiskers drawn at the lowest and highest points within 1.5 interquartile range of the lower and upper quartiles, respectively. Source data are provided as a Source Data file.