A deep learning method for predicting metabolite–disease associations via graph neural network
Authors:
Top Cited Papers
- 12 July 2022
- journal article
- research article
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 23 (4)
Abstract
Metabolism is the process by which an organism continuously replaces old substances with new substances. It plays an important role in maintaining human life, body growth and reproduction. More and more researchers have shown that the concentrations of some metabolites in patients are different from those in healthy people. Traditional biological experiments can test some hypotheses and verify their relationships but usually take a considerable amount of time and money. Therefore, it is urgent to develop a new computational method to identify the relationships between metabolites and diseases. In this work, we present a new deep learning algorithm named as graph convolutional network with graph attention network (GCNAT) to predict the potential associations of disease-related metabolites. First, we construct a heterogeneous network based on known metabolite–disease associations, metabolite–metabolite similarities and disease–disease similarities. Metabolite and disease features are encoded and learned through the graph convolutional neural network. Then, a graph attention layer is used to combine the embeddings of multiple convolutional layers, and the corresponding attention coefficients are calculated to assign different weights to the embeddings of each layer. Further, the prediction result is obtained by decoding and scoring the final synthetic embeddings. Finally, GCNAT achieves a reliable area under the receiver operating characteristic curve of 0.95 and the precision-recall curve of 0.405, which are better than the results of existing five state-of-the-art predictive methods in 5-fold cross-validation, and the case studies show that the metabolite–disease correlations predicted by our method can be successfully demonstrated by relevant experiments. We hope that GCNAT could be a useful biomedical research tool for predicting potential metabolite–disease associations in the future.Keywords
Funding Information
- National Natural Science Foundation of China (11805091)
- Foundation of Education Department of Liaoning Province (LJKZ0280)
This publication has 43 references indexed in Scilit:
- Capillary electrophoresis‐mass spectrometry‐based metabolome analysis of serum and saliva from neurodegenerative dementia patientsElectrophoresis, 2013
- Metabonomic Understanding of Probiotic Effects in Humans With Irritable Bowel SyndromeJournal of Clinical Gastroenterology, 2011
- Inferring the human microRNA functional similarity and functional network based on microRNA-associated diseasesBioinformatics, 2010
- Associating Genes and Protein Complexes with Disease via Network PropagationPLOS Computational Biology, 2010
- Coffee and cardiovascular disease: In vitro, cellular, animal, and human studiesPharmacological Research, 2007
- Modulation of cytokine secretion by pentacyclic triterpenes from olive pomace oil in human mononuclear cellsCytokine, 2006
- Effect of Pravastatin on Rate of Kidney Function Loss in People With or at Risk for Coronary DiseaseCirculation, 2005
- Different effect induced by treatment with several statins on monocyte tissue factor expression in hypercholesterolemic subjectsClinical and Experimental Medicine, 2003
- Remarkable increase in the concentration of 8‐hydroxyguanosine in cerebrospinal fluid from patients with Alzheimer's diseaseJournal of Neuroscience Research, 2002
- Plasma deoxycholic acid concentration is elevated in postmenopausal women with newly diagnosed breast cancerEuropean Journal of Clinical Nutrition, 2002