Fig. 3

Distribution of log-ranked P values using a bootstrap resampling technique. We compared a prognosis model with miR-eQTLs (blue) to one without miR-eQTLs (red) using a bootstrap resampling technique. This procedure was repeated 10,000 times. The distribution of log-rank P values was significantly different between the prediction models with and without miR-eQTLs (P < 0.01, Welch’s t test)