1: Category of the pairs of complexes. 2: number of pairs. 3: number of distinct SCOP families for A/A′ domains. 4: number of distinct SCOP families for B/C domains.
Each line corresponds to the best result (classifier) obtained for each source as we used the training and test data from 1 up to 100 features after SVD processing and the classification techniques Na
Cell level feature classification confusion matrix. Bold values indicate agreement between the classifier and the true class. Overall classification accuracy is 77.9%. The number of proteins in each c
