Figure 2.

(A) Venn diagram of the selected 484 features under each feature selection technique. It can be seen that a large number of features selected under LASSO and SVM are not common to the features selected under other techniques. On the other hand, most of the features selected through F-measure and Information gain are among those selected under other three techniques. (B) Performance metrics for SVM with RBF kernel with 484 selected GPC-0123 features under each feature selection technique. It can be observed that values of performance metrics are higher for RF-based selected 484 GPC-0123 features. (C) Distribution of length of sequences for each family of HSPs.