Prediction of Drug Targets for Specific Diseases Leveraging Gene Perturbation Data: A Machine Learning Approach.

Kai Zhao,Hon-Cheong So,Yujia Shi

doi:10.3390/pharmaceutics14020234

Abstract

Identification of the correct targets is a key element for successful drug development. However, there are limited approaches for predicting drug targets for specific diseases using omics data, and few have leveraged expression profiles from gene perturbations. We present a novel computational approach for drug target discovery based on machine learning (ML) models. ML models are first trained on drug-induced expression profiles with outcomes defined as whether the drug treats the studied disease. The goal is to “learn” the expression patterns associated with treatment. Then, the fitted ML models were applied to expression profiles from gene perturbations (overexpression (OE)/knockdown (KD)). We prioritized targets based on predicted probabilities from the ML model, which reflects treatment potential. The methodology was applied to predict targets for hypertension, diabetes mellitus (DM), rheumatoid arthritis (RA), and schizophrenia (SCZ). We validated our approach by evaluating whether the identified targets may ‘re-discover’ known drug targets from an external database (OpenTargets). Indeed, we found evidence of significant enrichment across all diseases under study. A further literature search revealed that many candidates were supported by previous studies. For example, we predicted PSMB8 inhibition to be associated with the treatment of RA, which was supported by a study showing that PSMB8 inhibitors (PR-957) ameliorated experimental RA in mice. In conclusion, we propose a new ML approach to integrate the expression profiles from drugs and gene perturbations and validated the framework. Our approach is flexible and may provide an independent source of information when prioritizing drug targets.

Highlights

We present a novel computational target discovery approach based on machine learning (ML) models to expression profiles induced by genetic perturbation
ML models were first trained on drug expression profiles to learn the expression patterns associated with treatment of a disease
It should be noted that the predictive performance of different ML methods is not the major focus of this study; our main objective is to uncover new disease drug targets and to validate our proposed approach by testing for its ability to ‘re-discover’ known targets

Summary

Introduction

Drug discovery involves a series of steps: target identification, target validation, lead identification, lead optimization, clinical trials, and introduction of the new drug to the market [1]. The speed of new drug development has been slower than anticipated, despite increasing investment [2]. It is estimated that the cost of developing a new drug is ≈USD 2.6 billion [3]. One of the main reasons for the enormous cost of drug discovery is the high failure rate. The success of drug development largely depends on the validity of targets. Most drugs fail to complete the development process due to a lack of efficacy, and Pharmaceutics 2022, 14, 234.

Objectives

Methods

Results

Discussion

Conclusion