RCDR: A Recommender Based Method for Computational Drug Repurposing

Abstract

Computational Drug repurposing is the problem of finding new uses for known drugs. To achieve this goal, a significant number of computational methods have been proposed, which can be categorized as Network-based and Non-network-based methods. Since network-based methods have a lot of advantages, this problem can be modelled as a network-based recommendation system. In this paper, we propose an effective approach, RCDR (Recommender Based Computational Drug Repurposing), to prioritize candidate drugs for diseases. Initially, we use drug and disease similarities to build a new drug-disease score matrix. Then, we adopt a collaborative filtering model to recommend which disease can be treated by the new drug. The experiment results show that RCDR proposes well performance compared with other state-of-the-art approaches.