Figure 2. Accuracy of VALORATE.

(A) Comparison of the exact log-rank distribution for a simulated dataset with the distribution estimated by VALORATE using ss = 100,000. The simulation was estimated by n = 100 subjects, d = 10 events, and n₁ = 7 mutations, which generates 16,007,560,800 combinations. (B) Densities in logarithm base 10 to show details in low-density regions (above 3). (C) Comparison of the p-values estimated by VALORATE and those estimated by ExaLT for the GBM dataset from Vandin et al.²⁷ shown in Supplementary Material (TableR.txt file in https://github.com/fvandin/ExaLT). p-values are shown in negative of the logarithm base 10.