Figure 3From: A Regression-based K nearest neighbor algorithm for gene function prediction from heterogeneous dataSensitivity at 50% accuracy based on functional classes. The methods used for comparison include svm.exp, loess.1000, glm.1111 and svm.comb. The X axis shows the index of the functional class. For descriptions of the classes, refer to Table 1. The Y axis shows the sensitivity of each class at 50% accuracy.Back to article page