Table of Contents Author Guidelines Submit a Manuscript
Computational and Mathematical Methods in Medicine
Volume 2015, Article ID 275045, 9 pages
http://dx.doi.org/10.1155/2015/275045
Research Article

Matrix Factorization-Based Prediction of Novel Drug Indications by Integrating Genomic Space

1Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China
2Beijing University of Chinese Medicine, Beijing 100029, China

Received 10 December 2014; Accepted 6 May 2015

Academic Editor: Francesco Pappalardo

Copyright © 2015 Wen Dai et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

There has been rising interest in the discovery of novel drug indications because of high costs in introducing new drugs. Many computational techniques have been proposed to detect potential drug-disease associations based on the creation of explicit profiles of drugs and diseases, while seldom research takes advantage of the immense accumulation of interaction data. In this work, we propose a matrix factorization model based on known drug-disease associations to predict novel drug indications. In addition, genomic space is also integrated into our framework. The introduction of genomic space, which includes drug-gene interactions, disease-gene interactions, and gene-gene interactions, is aimed at providing molecular biological information for prediction of drug-disease associations. The rationality lies in our belief that association between drug and disease has its evidence in the interactome network of genes. Experiments show that the integration of genomic space is indeed effective. Drugs, diseases, and genes are described with feature vectors of the same dimension, which are retrieved from the interaction data. Then a matrix factorization model is set up to quantify the association between drugs and diseases. Finally, we use the matrix factorization model to predict novel indications for drugs.