Fig. 1From: A deep learning approach to predict inter-omics interactions in multi-layer networksOverview of DIDL method framework. Rows and columns of \(R_{12}\) are fed to an encoder which consists of two DNNs. The outputs of the encoder are latent features of the biomolecules. Finally, these latent features are transformed into a tensor factorization predictor and its output indicates the probability of interaction between heterogeneous biomoleculesBack to article page