Extended Data Figure 1 |. Selection criteria for histone mutations to include in high-throughput libraries.
a, Distributions of histone mutations (tumor mutational burden ≤ 10) across the four canonical histones were fit to a Poisson distribution (red line) to identify sites of high mutational frequency for inclusion in the nucleosome library. Mutation values for each histone in the 99th percentile of the Poisson distribution are marked in red. b, The set of amino acid mutations observed at each histone site was assessed for overrepresentation of specific histone mutations by calculating the probability of selecting the observed number of specific mutations randomly, adjusting for codon bias. Probabilities were then adjusted for false discovery rates by the Benjamini-Hochberg method. FDR-adjusted p-value < 0.03 are shown in red.