MolDesign: Intelligent System For High-Performance Material Discovery

Problem Statement

Many researches on plant-based and chemical-based drug discoveries are going on for centuries, and it takes many years to make it into safe drugs and bring it into the market. The commercialization pathway is long and huge amount of funding needed. Failure rates using the traditional method is high when researchers need to deal and test with millions of molecules. Ensuring drug safety is one of the main challenges in the drug discovery process. Interpreting information on the known effects of drugs and predicting their side effects are complex tasks. In short, understanding the efficacy and safety of the new medicines takes lots of money and time.

Solution Statement

We all know that integrating computational analysis extraction leads to practical applications such as Machine Learning for Bioinformatics during the drug discovery process. Machine Learning could make in silico analysis more relevant since it can be a complementary technique that will lead to pre-clinical trials, enhancing competitiveness in drug development. Machine Learning for Bioinformatics can cut costs by predicting the efficacy of the new drug first before rigour of clinical trials and other tests.

Over many years, UM researchers have been collecting extensive and diverse data from various sources to enhance their set of Machine Learning for Bioinformatics automation protocol as part of research and client services. What has helped them is to automate repetitive data processing and analysis tasks apply at several steps during early drug discovery to:

• Predict target structure
• Identify and optimise “hits.”
• Explore the biological activity of new ligands
• Design models that predict the pharmacokinetic and toxicological properties of the drug candidate