Identifying Stage II Colorectal Cancer Recurrence Associated Genes by Microarray Meta-Analysis and Building Predictive Models with Machine Learning Algorithms

Figure 7

(a) Violin plot of the predicted recurrence risk scores in the test set. Patients who did not experience recurrence are shown in green, and patients who experienced recurrence are shown in red. The Mann–Whitney test was adopted to examine whether predicted recurrence risk scores had statistical significance between groups. (b) Kaplan–Meier survival curve of the random survival forest model prediction results in the test set. Patients who had higher recurrence risk scores than the cut-off value were predicted to be the high risk group (red curve), and patients who had lower recurrence risk scores than the cut-off value were predicted to be the low risk group accordingly (green curve). The log-rank test was performed to find whether survival difference between groups was statistically significant.

(a)

(b)