Fig. 3: Multiple feature separation by condition and sex.

Left Panels. Analyses of differences in effect size separation between control condition and test condition for each sex for all features. A CNC cohort compared to D cohort. B CNC cohort compared to H cohort. C CNC cohort compared to B cohort. D D cohort compared to H cohort. The solid identity line is where the effect size for a feature is equal in females and males. Parallel dotted lines below the identity line reflect a 5% increment increase in feature separability in females, and dotted lines above the identity line reflect a 5% increase in feature separability in males. Gray, blue, red, and purple markers indicate features that are significant in neither sex, females only, males only, or both sexes, respectively. Right Panels. Separability of features by feature type within and across sex. Red and blue boxplots are comprised of the effect sizes from features that are significant in males and females, respectively. Each individual boxplot was evaluated with a one-sided Wilcoxon signed rank test to evaluate if the aggregate effect size of significant features is equal than 0, with significance annotated directly above the boxplots (*p < 0.05, **p < 0.01, ***p < 0.001). For each feature type, a paired, two-sided Wilcoxon signed rank test was used to evaluate if the effect sizes from significant features are equal between males and females (*p < 0.05, **p < 0.01, ***p < 0.001)