Fig. 5

DCA, CICA and machine learning modeling. A DCA based on the five hub genes. The x-axis represented the threshold probability and the y-axis represented the net benefit. The gray line indicated a perfect prediction model, while the dash line represented a noneffective prediction model. The prevalence rate of CHD (0.07) was also marked in the figure. B CICA based on the five hub genes. The number of high-risk patients and the number of high-risk patients with events were drawn with solid lines and dash lines to represent different threshold probabilities, respectively. C–N Performance of the machine learning models constructed by LightGBM. ROC, PR, lift chart and KS plot of the methylation model (C–F), expression model (G–J) and combination model (K–N). O–R Risk plot of the validation set in the methylation model (O), expression model (P), combination model (Q) and FRS model (R)