Drug Similarity Integration Through Attentive Multi-view Graph Auto-Encoders

By Tengfei Ma, Cao Xiao, Jiayu Zhou, Fei Wang. 2018

189 citations recorded by the source corpus.

Drug similarity has been studied to support downstream clinical tasks such as inferring novel properties of drugs (e.g. side effects, indications, interactions) from known properties. The growing availability of new types of drug features brings the opportunity of learning a more comprehensive and accurate drug similarity that represents the full spectrum of underlying drug relations. However, it is challenging to integrate these heterogeneous, noisy, nonlinear-related information to learn accurate similarity measures especially when labels are scarce. Moreover, there is a trade-off between accuracy and interpretability. In this paper, we propose to learn accurate and interpretable similarity measures from multiple types of drug features. In particular, we model the integration using multi-view graph auto-encoders, and add attentive mechanism to determine the weights for each view with respect to corresponding tasks and features for better interpretability. Our model has flexible design for both semi-supervised and unsupervised settings. Experimental results demonstrated significant predictive accuracy improvement. Case studies also showed better model capacity (e.g. embed node features) and interpretability.

Subfields: Computational Theory and Mathematics, Molecular Biology, Pharmacology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

In Pharmacology

Citations in the corpus

Unsupervised Inductive Whole-Graph Embedding by Preserving Graph Proximity (2019, background): “For example, in drug-drug interaction detection Ma et al. (2018), a domain-specific metric to encode compound chemical structure can be used to compute the similarities between chemical graphs.”
Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity (2019, neutral): “For example, in drug-drug interaction detection [Ma et al., 2018], a domain-specific metric to encode compound chemical structure can be used to compute the similarities between chemical graphs.”
Artificial intelligence in cancer target identification and drug discovery (2022, background): “For example, Ma et al. 172 proposed a novel graph autoencoders model (GAE) to learn accurate and interpretable drug similarity measures from multiple types of drug properties.”
GAMENet: Graph Augmented MEmory Networks for Recommending Medication Combination (2018, methodology): “For example, [17] models each drug as a node and DDIs as node labels in the drug association network and extended the GCN to embed multi-view drug features and edges.”
Machine Learning for Integrating Data in Biology and Medicine: Principles, Practice, and Opportunities (2018, background): “all, a given drug pair manifests clinically within a patient population [45, 302, 303].”
SimGNN: A Neural Network Approach to Fast Graph Similarity Computation (2018, background): “A great amount of graphbased applications have been tackled by neural network basedmethods, most of which are framed as node-level prediction tasks [27, 37].”
Graph Representation Learning in Biomedicine and Healthcare (2021, unknown): “…been applied on drug graphs, with chemical structures and side effects as features, to generate interpretable predictions of drug–drug interactions 159 . by networks, networks of EHRs can be fused with biomedical networks, thus enabling graph representation learning to make predictions on…”
Graph embedding on biomedical networks: methods, applications and evaluations (2019, result): “For predicting PPIs, Laplacian and SVD are commonly adopted (Zhu et al., “main” — 2019/6/13 — page 4 — #4 Method Catergory Embdding Methods Link Prediction Tasks Node Classification Task drug-disease association prediction drug-drug interaction prediction protein-protein interaction prediction medic”
Graph convolutional networks for computational drug development and discovery (2019, background): “[42] proposed a multi-view drug similarity integration framework using graph auto-encoders with attention mechanism.”