Dr. Guowei Wei is developing artificial intelligence technology to make drug discovery faster and more affordable.
AI is capable of massively transforming several industries like retail, consumer internet, financial services, and healthcare, but its potential for drug discovery is especially exciting. Unfortunately, a number of obstacles stand in the way of advancement, including the high dimensionalities of drug candidates’ chemical space and complexity of protein-ligand interactions. Currently, computers do not have sufficient computing power to discover new drugs.
Wei and his team are addressing these challenges through mathematics by focusing on the reduction of the geometric complexity for computers.
Wei is a professor in the Department of Mathematics at Michigan State University and a researcher in a relatively young field known as mathematical molecular bioscience and biophysics.
“I have great satisfaction in working closely with my students and postdocs and interacting with my collaborators and industrial partners from Pfizer and Bristol-Myers Squibb,” said Wei. “I enjoy the rapid advance of new mathematical and deep learning strategies for biomedicine and am excited to see how advanced mathematics will further work to tackle challenge biological problems.”
On average, a new drug takes over 10 years and $2.6 billion to bring to market. It is therefore unprofitable to invest in drug discovery for rare medical ailments and many existing, effective drugs are too expensive for the average American to reasonably afford. In addition, effective treatments for many common medical conditions, such as Alzheimer’s disease, have not yet been discovered.
A promising alternative to current drug discovery that would ultimately nullify these issues is artificial intelligence (AI).
Wei and his team have introduced differential geometry, algebraic topology and graph theory to obtain high-level abstractions of protein-drug interactions. This significantly enhances the ability of computers to handle the high dimensionality, structural complexity and excessively large datasets involved in new drug discovery through AI. In the past three years, Wei’s team has been a top competitor in D3R Grand Challenges, a worldwide annual competition series in computer-aided drug design.
Wei worked with the MSU Innovation Center to obtain his US patent. “The Innovation Center helps us to understand the importance and potential of commercialization,” said Wei.