GraphDTA: Predicting drug–target binding affinity with graph neural networks

Abstract

Background: While the development of new drugs is costly, time consuming, and often accompanied with safety issues, drug repurposing, where old drugs with established safety are used for medical conditions other than originally developed, is an attractive alternative. Then, how the old drugs work on new targets becomes a crucial part of drug repurposing and gains much of interest. Several statistical and machine learning models have been proposed to estimate drug-target binding affinity and deep learning approaches have been shown to be among state-of-the-art methods. However, drugs and targets in these models have been commonly represented in 1D strings, regardless the fact that molecules are by nature formed by the chemical bonding of atoms. Method: In this work, we propose GraphDTA to capture the structural information of drugs, possibly enhancing the predictive power of the affinity. In particular, unlike competing methods, drugs are represented as graphs and graph convolutional networks are used to learn drug-target binding affinity. We trial our method on two benchmark datasets of drug-target binding affinity and compare the performance with state-of-the-art models in the research field. Results: The results show that our proposed method can not only predict the affinity better than non-deep learning models, but also outperform competing deep learning approaches. This demonstrates the practical advantages of graph-based representation for molecules in providing accurate prediction of drug-target binding affinity. The application may also include any recommendation systems where either or both of the user-and product-like sides can be represented in graphs. Availability and implementation: The proposed models are implemented in Python. Related data, pre-trained models, and source code are publicly available at https://github.com/thinng/GraphDTA.

Graph convolutional networks for drug response prediction

“…CNNs have recently achieved success in computer vision [46], [47] and natural language processing [48], [49], which motivates the use of convolutional neural networks to graph structures. Similar to the use of CNN with image, CNN in the graph also has two main layers including convolutional layer to learn receptive fields in graphs whose data points are not arranged as Euclidean grids and pooling layer to down-sample a graph [35]. Graph convolutional network (GCN) is well-fitting for the drug response problem because the drug molecular itself is represented in the form of a graph.…”

“…Graph convolutional network (GCN) is well-fitting for the drug response problem because the drug molecular itself is represented in the form of a graph. In order to evaluate the effectiveness of graph-based models, we investigate several graph convolutional models, including GCN [43], GAT [44], GIN [45] and combined GAT-GCN architecture [35]. The details of each GCN architecture are described as follows.…”

“…A combination of GAT [44] and GCN [43] is also proposed to learn graph feature [35]. At first, the input graph is passed to GAT layers.…”

“…Graph convolutional networks are able to learn representations of compound structures represented as molecular graphs [34]. For example, in GraphDTA [35], the drugs are presented as graphs where the edges are the bonding of atoms and the model achieves the best performance compared to other deep learning-based methods which represent drugs as strings in the task of drug-target binding affinity prediction. However, the graph neural network has not been employed yet [34] for the drug response prediction problem.…”

Uncovering new drug properties in target-based drug-drug similarity networks

“…Some approaches 31 build drug-drug similarity networks, where the similarity is defined based on 32 transcriptional responses [22,23]. These repositioning approaches analyze the network 33 parameters and the node centrality distributions in either drug-drug or drug-target 34 networks, using statistical analysis [10,11,24,25] and machine learning (i.e., graph 35 convolutional networks) [26][27][28][29] to link certain drugs to new pharmacological properties. 36 However, conventional statistics can be misleading when used to predict extreme 37 centrality values, such as degree and betweenness (which particularly indicate 38 nodes/drugs with a high potential for repositioning) [30].…”