Skip to main content


Fig. 2 | BMC Systems Biology

Fig. 2

From: A unified solution for different scenarios of predicting drug-target interactions via triple matrix factorization

Fig. 2

Illustration of triple matrix factorization over four predicting scenarios. The big rounded rectangle indicates the triple matrix factorization. Fd, x is the feature vector of the new drug dx and Ft, y is the feature vector of the new protein ty. The predicted confidence scores for S1, S2, S3 and S4 are denoted by \( \tilde{\mathbf{A}} \),\( {\tilde{\mathbf{D}}}_{\mathbf{A}}^x \),\( {\mathbf{T}}_{\mathbf{A}}^y \) and \( {\tilde{\tilde{\mathbf{A}}}}_{x,y} \) respectively

Back to article page