Fig. 5.

Accuracy of the classification using three different classification methods. A ROC curve and AUC parameter were calculated for: support vector machines (SVM; blue line), linear discriminate analysis LDA (red line), and random forest (RF; green line). Results were obtained based on B-lymphocyte cell lines (n = 20 from CEU and CHB). The ROC curve was created by plotting the true positive fraction against the false positive fraction at various threshold settings