A two-tiered unsupervised clustering approach for drug repositioning through heterogeneous data integration

Abstract

BackgroundDrug repositioning is the process of identifying new uses for existing drugs. Computational drug repositioning methods can reduce the time, costs and risks of drug development by automating the analysis of the relationships in pharmacology networks. Pharmacology networks are large and heterogeneous. Clustering drugs into small groups can simplify large pharmacology networks, these subgroups can also be used as a starting point for repositioning drugs. In this paper, we propose a two-tiered drug-centric unsupervised clustering approach for drug repositioning, integrating heterogeneous drug data profiles: drug-chemical, drug-disease, drug-gene, drug-protein and drug-side effect relationships.ResultsThe proposed drug repositioning approach is threefold; (i) clustering drugs based on their homogeneous profiles using the Growing Self Organizing Map (GSOM); (ii) clustering drugs based on drug-drug relation matrices based on the previous step, considering three state-of-the-art graph clustering methods; and (iii) inferring drug repositioning candidates and assigning a confidence value for each identified candidate. In this paper, we compare our two-tiered clustering approach against two existing heterogeneous data integration approaches with reference to the Anatomical Therapeutic Chemical (ATC) classification, using GSOM. Our approach yields Normalized Mutual Information (NMI) and Standardized Mutual Information (SMI) of 0.66 and 36.11, respectively, while the two existing methods yield NMI of 0.60 and 0.64 and SMI of 22.26 and 33.59. Moreover, the two existing approaches failed to produce useful cluster separations when using graph clustering algorithms while our approach is able to identify useful clusters for drug repositioning. Furthermore, we provide clinical evidence for four predicted results (Chlorthalidone, Indomethacin, Metformin and Thioridazine) to support that our proposed approach can be reliably used to infer ATC code and drug repositioning.ConclusionThe proposed two-tiered unsupervised clustering approach is suitable for drug clustering and enables heterogeneous data integration. It also enables identifying reliable repositioning drug candidates with reference to ATC therapeutic classification. The repositioning drug candidates identified consistently by multiple clustering algorithms and with high confidence have a higher possibility of being effective repositioning candidates.Electronic supplementary materialThe online version of this article (10.1186/s12859-018-2123-4) contains supplementary material, which is available to authorized users.

Artificial Intelligence in Drug Discovery: A Comprehensive Review of Data-driven and Machine Learning Approaches

“…Interactions or relationships between various types of node data such as drugs, diseases, genes, and proteins, as well as each characteristic information of drug and disease, are considered in heterogeneous networks. These interactions contribute to identifying drug repositioning candidates from various perspectives [245]. Thus, most studies first constructed a heterogeneous network, and then a network-based algorithm was applied.…”

“…[258] Kim et al [259] constructed a drug-disease association prediction model with five ML algorithms considering both linear and nonlinear algorithms, using similarities as features representing drug-disease pairs. Besides supervised models, there are also studies using unsupervised algorithms [245]. Hameed et al used four clustering algorithms on a drug network to predict drug ATC classes.…”

Unsupervised Learning for Large Scale Data: The ATHLOS Project

“…Nevertheless, techniques and methods that fall into the category of unsupervised learning have shown encouraging results and are able to overcome some of the difficulties of vast, heterogeneous data (Casolla, et al 2019) (Hameed, et al 2018) (Ma, et al 2017) (Xiang, et al 2018). In this study, we focus on the area of unsupervised learning, presenting a complete methodological procedure that utilizes recent advances in the field.…”

Breaking the paradigm: Dr Insight empowers signature-free, enhanced drug repurposing

“…Large-scale drug-perturbed gene expression datasets, such as Connectivity Map (CMap) (Lamb et al , 2006; Subramanian et al , 2017), provide unprecedented opportunities for prioritizing treatments based on the associations between disease state and chemical intervention. Numerical computational approaches have been developed, taking full advantage of these high-throughput resources, for in silico prediction of disease–drug connectivity and drug–drug connectivity (El-Hachem et al , 2017; Hameed et al , 2018; Iorio et al , 2010, 2013; Lee et al , 2016a; Peyvandipour et al , 2018; Sirota et al , 2011). Notable successes have been achieved using CMap and its variants to uncover novel therapeutic redirections of existing drugs to treat various types of diseases, including obesity (Lee et al , 2016b; Liu et al , 2015a), neurodegenerative diseases (Sandor et al , 2017; Siavelis et al , 2016), gastrointestinal and liver diseases (Hicks et al , 2017), stroke and sepsis (Chen et al , 2015b) and cancers (Hsieh et al , 2016; Liu et al , 2015b; Xiang et al , 2016; Zhao et al , 2016).…”