Figure 3
From: Predicting mortality among ischemic stroke patients using pathways-derived polygenic risk scores
Kaplan–Meier plot of the two groups created by the cutpoint for PRS. All the pathway-specific PRSs for 3-year mortality identified by univariate CoxPH were dichotomized by the corresponding cutpoint, and Kaplan–Meier analysis for each binary PRS was conducted. We simulated the null distribution using the conditional Monte Carlo method and compared it with the exact distribution for the log-rank statistic to get the lower bound of the p-value for each pathway-specific PRS. P value derived from the Log-rank test was labeled.
