The accuracies of prediction models constructed using our algorithm.

aKnown interactions for building classifier model, which were collected till Jan, 2011.bThe 5-CV performance of statistical learning methods can be measured by the quantity of true positives (TP), true negatives (TN), false positives (FP) and false negatives (FN). Precision [PRE = TP/(TP+FP)] is a measure of the accuracy provided that a specific class has been predicted. Accuracy [ACC = (TP+TN)/(TP+TN+FP+FN)] is another frequently used index for the overall classification performance, but it may be misleading as a result of highly unbalanced class distribution of used datasets. Sensitivity [SE = TP/(TP+FN)] and specificity [SP = TN/(TN+FP)] can assess a model's ability to correctly identify TP and TN, respectively, while they are usually interpreted in combination with each other.