Drug Repositioning for Schizophrenia and Depression/Anxiety Disorders: A Machine Learning Approach Leveraging Expression Data

Abstract

Development of new medications is a lengthy and costly process, and drug repositioning might help to shorten the development cycle. We present a machine learning (ML) workflow to drug discovery or repositioning by predicting indication for a particular disease based on drug expression profiles, with a focus on applications in psychiatry. Drugs that are not originally indicated for the disease but with high predicted probabilities serve as candidates for repurposing. This approach is widely applicable to any chemicals or drugs with expression profiles measured, even if drug targets are unknown. It is also highly flexible as virtually any supervised learning algorithms can be used. We employed the ML approach to identify repositioning opportunities for schizophrenia as well as depression and anxiety disorders. We applied various state-of-the-art ML approaches, including deep neural networks (DNN), support vector machines (SVM), elastic net regression, random forest and gradient boosted trees. The predictive performance of the five approaches in cross-validation did not differ substantially, with SVM slightly outperforming the others. However, other methods also reveal literature-supported repositioning candidates of different mechanisms of actions. As a further validation, we showed that the repositioning hits are enriched for psychiatric medications considered in clinical trials. We also examined the correlation between predicted probabilities of treatment potential and the number of related research articles, and found significant correlations for all methods, especially DNN. Finally, we propose that ML may provide a new avenue to exploring drug mechanisms via examining the variable importance of gene features.

Examination of retinal vascular trajectory in schizophrenia and bipolar disorder

“…Our secondary aim was to examine the utility of machine‐learning techniques to differentiate patients from HV using retinal vessel trajectory as the input variable. Several studies have used the machine‐learning approach to classify SCZ/BD and HV using neuroimaging/electrophysiologic/cognitive data as input variables . A recent study reported prediction accuracy of 76% to differentiate SCZ from HV using a multisite machine‐learning analysis .…”

Turning genome-wide association study findings into opportunities for drug repositioning

“…ML approaches such as deep learning and other methods hold great promise for accelerating drug discovery/repositioning, as they may be able to discover and predict with higher accuracy the complex patterns and relationships between genes, drugs and diseases [130] , [131] , [132] , [133] . For example, ML methods may be used to capture complex relationships between drug transcriptome and the drug’s treatment potential for specific diseases [134] . For diseases with high heterogeneity, a drug may only be useful for a subgroup of patients [135] .…”

Drug targets for COVID-19 therapeutics: Ongoing global efforts

“…Huge data generated by High-throughput Next Gen Sequencing (NGS) from numerous patients when combined with disease characteristics and treatment options can lead to the identification of new disease biomarkers and drug targets(Stupnikov et al 2018;Zai et al 2018). AI-driven supervised machine learning algorithms can implement multiomics and multitask learning to facilitate drug response elicited by engagement of multiple drug targets(Nascimento et al 2019;Nath et al 2018;Saberian et al 2019;Zhao and So 2019).…”