Application of Computational Models Using Machine Learning Methods to Predict Drug Toxicity and Advance Drug Development

Drug development is a complex process that is costly, time-consuming, and has a relatively low success rate. Despite these challenges, this process is essential to better understand the specific pathways and targets involved in disease etiology and ultimately develop effective interventions essential for the improvement of human health. Toxicity testing is a critical step used to determine the adverse effect potential of the drug in development, and it is increasingly relying on the in silico modeling approach. For instance, in the cases of drug-induced liver injury (DILI) and cardiotoxicity (DICT) current in vivo toxicological testing is insufficient to comprehensively assess the hepatotoxic and cardiotoxic potential of compounds, thereby presenting an urgent need for alternative prediction strategies. In the adjacent areas of drug development, drug repurposing became an appealing method to address the Coronavirus 2019 (COVID-19) pandemic because of the low cost and efficiency. Compounds that exhibited anti-SARS-CoV-2 activity were found to correlate with other biological activities (e.g., human ether-a-go-go-related gene (hERG), phospholipidosis (PLD), and cytotoxicity screens), so these compounds need to be evaluated for their toxicity potential especially in terms of cardiotoxicity via hERG inhibition. Lastly, there has been considerable interest in recent years in the development of small molecule protein kinase inhibitors for the treatment of several diseases. While a number of protein kinase inhibitors have been FDA approved, there remains a need to develop effective computational models to predict protein kinase inhibition activity of small molecules. Collectively, computational approaches using machine learning algorithms can significantly improve the drug development and toxicity evaluation of small molecules that target unmet medical needs across diverse therapeutic areas.

Author

• Ngan, Deborah Kim

• cardiotoxicity
• COVID-19
• drug development
• hepatotoxicity
• kinase inhibitor
• toxicity

Identifier

hdl:1920/14485

OAI identifier oai:identifier

oai:MARS:1920/14485