ci9b00628_si_001.pdf (367.89 kB)
Graph Convolutional Neural Networks for Predicting Drug-Target Interactions
journal contribution
posted on 2019-10-16, 18:35 authored by Wen Torng, Russ B. AltmanAccurate determination of target-ligand
interactions is crucial
in the drug discovery process. In this paper, we propose a graph-convolutional
(Graph-CNN) framework for predicting protein-ligand interactions.
First, we built an unsupervised graph-autoencoder to learn fixed-size
representations of protein pockets from a set of representative druggable
protein binding sites. Second, we trained two Graph-CNNs to automatically
extract features from pocket graphs and 2D ligand graphs, respectively,
driven by binding classification labels. We demonstrate that graph-autoencoders
can learn fixed-size representations for protein pockets of varying
sizes and the Graph-CNN framework can effectively capture protein-ligand
binding interactions without relying on target-ligand complexes. Across
several metrics, Graph-CNNs achieved better or comparable performance
to 3DCNN ligand-scoring, AutoDock Vina, RF-Score, and NNScore on common
virtual screening benchmark data sets. Visualization of key pocket
residues and ligand atoms contributing to the classification decisions
confirms that our networks are able to detect important interface
residues and ligand atoms within the pockets and ligands, respectively.
History
Usage metrics
Categories
Keywords
screening benchmark data setsGraph Convolutional Neural Networksbinding classification labelsrepresentative druggable protein binding sitesprotein pocketsprotein-ligand binding interactions3 DCNN ligand-scoringgraph-autoencoder2 D ligand graphsfixed-size representationsdrug discovery processresidueframeworkGraph-CNNtarget-ligandligand atoms
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC