NEDD: a network embedding based method for predicting drug-disease associations

Abstract

Background Drug discovery is known for the large amount of money and time it consumes and the high risk it takes. Drug repositioning has, therefore, become a popular approach to save time and cost by finding novel indications for approved drugs. In order to distinguish these novel indications accurately in a great many of latent associations between drugs and diseases, it is necessary to exploit abundant heterogeneous information about drugs and diseases. Results In this article, we propose a meta-path-based computational method called NEDD to predict novel associations between drugs and diseases using heterogeneous information. First, we construct a heterogeneous network as an undirected graph by integrating drug-drug similarity, disease-disease similarity, and known drug-disease associations. NEDD uses meta paths of different lengths to explicitly capture the indirect relationships, or high order proximity, within drugs and diseases, by which the low dimensional representation vectors of drugs and diseases are obtained. NEDD then uses a random forest classifier to predict novel associations between drugs and diseases. Conclusions The experiments on a gold standard dataset which contains 1933 validated drug–disease associations show that NEDD produces superior prediction results compared with the state-of-the-art approaches.

Multiplex-Heterogeneous Network Embedding for Drug Repositioning

“…So far, it focused on heterogeneous network embedding [3,25,38,38,37]. Zhou and colleagues focused on a drug-disease heterogeneous network [39]. The method called NEDD, applied meta paths of different lengths to explicitly capture the similarities within drugs and diseases, by which they optimize the embeddings of drugs and diseases.…”

“…In this article, Yang et al proposed HED to predict potential associations between drugs and diseases based on a drug-disease heterogeneous network. From the embeddings, similarly to [39], they trained an SVM binary classifier to predict new associations. Chen and colleagues introduced cross-network embedding to embed drugs, targets and diseases nodes using two heterogeneous networks, a drug-target and a drug-disease network [3].…”

Meta-Path Based Gene Ontology Profiles for Predicting Drug-Disease Associations

“…Zeng et al developed deepDR, a deep learning based method that integrates ten networks of drugs and diseases, for computational drug repositioning [9]. Based on a two-layer network of drug-drug and diseasedisease similarities, Zhou et al proposed a network embedding based method for predicting drug-disease associations (NEDD) [10]. Recently, Yu et al used drugdrug similarities based on five drug properties and diseasedisease similarities to develop a layer attention graph convolutional network (LAGCN) method for drug repositioning [11].…”

Cascade Deep Forest With Heterogeneous Similarity Measures for Drug–Target Interaction Prediction

“…Along with sustainable innovative developments of biological data, and high-speed improvements of machine learning technology in recent years, a variety of methods for computational drug repositioning have been put forward correspondingly and achieved some achievements in practical applications ( Lan et al, 2016 , 2020 ; Chen et al, 2021 ; Li et al, 2019 ; Liu et al, 2019 ; Zeng et al, 2019 ; Fahimian et al, 2020 ; Rauschenbach et al, 2020 ; Zhou et al, 2020 ; Jarada et al, 2021 ; Meng et al, 2021 ). Machine learning is a beneficial complement to ligand-based and structure-based methods.…”

“…RepCOOL recommends four novel drugs for the treatment of breast cancer at stage II, namely, paclitaxel, doxorubicin, tamoxifen, and trastuzumab. In addition, a network embedding based method for predicting drug–disease interactions (NEDD) is raised ( Zhou et al, 2020 ). Initially, through constructing a heterogeneous network and utilizing meta-paths of various lengths, NEDD accurately obtains the indirect associations between drugs and diseases or their strong proximity, thereby acquiring representation vectors of drugs and diseases with low dimensions.…”