Skip to main content
Fig. 3 | Journal of Cheminformatics

Fig. 3

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

Fig. 3

Visualization of sub-graph embeddings for the target local environment. The scatter plot was created by reducing the number of dimensions with the Principal Component Analysis (PCA) of five different subgraphs (represented by different maker shapes), each with several different sizes. The size is defined as the number of highly confident neighbors of a target node (N) increasing from 10 to 70 (shown in different colors)

Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us