Show simple item record

dc.contributor.authorHan, Peng
dc.contributor.authorShang, Shuo
dc.contributor.authorYang, Peng
dc.contributor.authorLiu, Yong
dc.contributor.authorZhao, Peilin
dc.contributor.authorZhou, Jiayu
dc.contributor.authorGao, Xin
dc.contributor.authorKalnis, Panos
dc.date.accessioned2019-09-17T06:30:16Z
dc.date.available2019-09-17T06:30:16Z
dc.date.issued2019-07-26
dc.identifier.citationHan, P., Yang, P., Zhao, P., Shang, S., Liu, Y., Zhou, J., … Kalnis, P. (2019). GCN-MF. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. doi:10.1145/3292500.3330912
dc.identifier.doi10.1145/3292500.3330912
dc.identifier.urihttp://hdl.handle.net/10754/656764
dc.description.abstractDiscovering disease-gene association is a fundamental and critical biomedical task, which assists biologists and physicians to discover pathogenic mechanism of syndromes. With various clinical biomarkers measuring the similarities among genes and disease phenotypes, network-based semi-supervised learning (NSSL) has been commonly utilized by these studies to address this class-imbalanced large-scale data issue. However, most existing NSSL approaches are based on linear models and suffer from two major limitations: 1) They implicitly consider a local-structure representation for each candidate; 2) They are unable to capture nonlinear associations between diseases and genes. In this paper, we propose a new framework for disease-gene association task by combining Graph Convolutional Network (GCN) and matrix factorization, named GCN-MF. With the help of GCN, we could capture nonlinear interactions and exploit measured similarities. Moreover, we define a margin control loss function to reduce the effect of sparsity. Empirical results demonstrate that the proposed deep learning algorithm outperforms all other state-of-the-art methods on most of metrics.
dc.publisherAssociation for Computing Machinery (ACM)
dc.relation.urlhttp://dl.acm.org/citation.cfm?doid=3292500.3330912
dc.rightsArchived with thanks to the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019
dc.subjectgraph convolutional networks
dc.subjectdeep learning
dc.subjectisease-gene associa-tion
dc.titleGCN-MF: Disease-gene association identification by graph convolutional networks and matrix factorization
dc.typeConference Paper
dc.contributor.departmentKing Abdullah University of Science and Technology
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputational Bioscience Research Center (CBRC)
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.conference.date2019-08-04 to 2019-08-08
dc.conference.name25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019
dc.conference.locationAnchorage, AK, USA
dc.eprint.versionPost-print
dc.contributor.institutionUniversity of Electronic Science and Technology of China, Inception Institute of Artificial Intelligence
dc.contributor.institutionCognitive Computing Lab, Baidu Research, USA
dc.contributor.institutionAlibaba-NTU Singapore Joint Research Institute, Nanyang Technological University
dc.contributor.institutionTencent AI Lab
dc.contributor.institutionMichigan State University
kaust.personHan, Peng
kaust.personGao, Xin
kaust.personKalnis, Panos
refterms.dateFOA2019-09-19T05:25:08Z
dc.date.published-online2019-07-26
dc.date.published-print2019


Files in this item

Thumbnail
Name:
kdd19-sigconf.pdf
Size:
799.8Kb
Format:
PDF
Description:
Accepted manuscript

This item appears in the following Collection(s)

Show simple item record