Skip to main content
Fig. 5 | Journal of Cheminformatics

Fig. 5

From: GraphDTI: A robust deep learning predictor of drug-target interactions from multiple heterogeneous data

Fig. 5

Separation of input features and output-layer embeddings in a low-dimensional space. The T-distributed Stochastic Neighbor Embedding (t-SNE) technique is applied to 500 positive (teal) and 500 negative (salmon) instances randomly selected from the PubChem BioAssay dataset. Dimensionality reduction is conducted for A 400-dimensional input feature vectors and B output-layer embeddings prior to the softmax activate function

Back to article page