Overall dependence and similarity between methods

aMI, mutual information. The mutual information between variables A and B (in this study, A and B represent two methods' prediction results) is calculated as MI(A,B) = H(A)+H(B)?H(A,B) where and [p(a) and p(a,b) are marginal and joint probability distributions, respectively]. The ten highest values are underlined.bJCs, Jaccard coefficients. The Jaccard coefficient between binary variables A and B (in this study, A and B represent two methods' prediction results) is calculated as JC(A,B) = p(1,1)/(1?p(0,0)) where p(a,b) is the joint probability distribution of A and B. The ten highest values are underlined.