Fig. 4: The elbow-point-based method offers one systematic way to determine the cutoff used for calling CNAs.
a The elbow method for determining the cutoff of |log2 copy ratio| was used in calling AS for individual cancer types (exemplified by esophagogastric cancer here). The AS for each patient with different calling cutoffs are shown in black curves. The mean value of all patients is shown in the red curve. The mean elbow point is shown with 95% confidence intervals, which are calculated using 1000-replicate bootstrapping. b The elbow-point values of the cutoff of |log2 copy ratio| in calculating AS in individual cancer types. The bars represent 95% confidence intervals of the elbow-point values calculated using a 1000-replicate bootstrapping. c Comparison of HRs using AS0.1 or ASEP in a multivariate Cox model with TMB (binned at the 80th percentile) and ICB drug class. The difference of mean HRs of AS0.1 and ASEP and paired Wilcoxon test P value are displayed. Wald P values for HRs of ASEP in individual cancer types are displayed at the right side of the plot. The upper and lower boundaries signify the first and third quartiles, correspondingly, while the central line denotes the median. Whiskers stretch to the most distant data points not classified as outliers (within 1.5 times the interquartile range), and outliers are illustrated as points above and below the box-and-whisker diagram. d Univariable Kaplan–Meier survival analysis and multivariable survival analysis using Cox proportional hazards regression of overall survival with AS calculated using cancer-type-specific elbow-point-based CNA calling cutoff (ASEP; binned at the 30th percentile), TMB (binned at the 80th percentile), and ICB drug class. Wald P values are displayed. Squares positioned at midpoints symbolize point estimates of HRs, and the accompanying bars indicate 95% confidence intervals. The data are from the Samstein et al.’s cohort1.
