MSU Study: AI Speeds Up Therapeutic Drug Discovery

Researchers use machine learning to predict how chemicals will influence gene expression and identify promising compounds for treating liver cancer and lung disease.

Mar. 18, 2026 at 5:14am

Got story updates? Submit your updates here. ›

A team of researchers led by Michigan State University has demonstrated a new approach using machine learning to speed up the process of discovering therapeutic drugs. By training an AI model on published data, they were able to predict how chemicals will influence gene expression and identify promising compounds for treating the most aggressive form of liver cancer and a chronic lung disease with no curative options. The findings resulted from a collaborative effort across multiple disciplines and institutions.

Why it matters

The new approach could lead to faster drug discovery for difficult-to-treat diseases like hepatocellular carcinoma (HCC), the third leading cause of cancer-related death worldwide, and idiopathic pulmonary fibrosis (IPF), a chronic lung disease with a median survival rate of just three years after diagnosis. Both diseases have an urgent need for new and more effective treatments.

The details

The researchers trained a "Gene expression profile Predictor on chemical Structures" (GPS) model on millions of experimental measurements to learn how chemicals influence gene expression. This allowed them to screen a large pool of compounds and identify promising candidates for further testing on cell lines and animal models. For HCC, they found two new compounds that reduced tumor size in mice. For IPF, they identified one repurposed drug and two new compounds that showed promise, testing them on samples of human lung tissue thanks to a collaboration with a clinical lung transplant program.

  • The study was recently published in the journal Cell in 2026.
  • The research has been a long-term collaborative effort spanning multiple years.

The players

Bin Chen

Associate professor at the College of Human Medicine in the departments of Pediatrics and Human Development and Pharmacology and Toxicology at Michigan State University, and one of the senior authors of the study.

Jiayu Zhou

Formerly at Michigan State University, now at the University of Michigan, and another senior author of the study.

Samuel So

The Lui Hac Minh Professor and contributing author from the Asian Liver Center at Stanford University.

Mei-Sze Chua

Senior research scientist at the Asian Liver Center at Stanford University and contributing author.

Xiaopeng Li

Associate professor in the Department of Pediatrics and Human Development in the College of Human Medicine at Michigan State University and another senior author.

Got photos? Submit your photos here. ›

What they’re saying

“Our previous efforts were limited to repurposing FDA-approved drugs. This new approach greatly expands the pool of novel compounds with potential therapeutic activity in HCC.”

— Mei-Sze Chua, Senior research scientist

“With the incidence of HCC continuing to increase in the USA, novel and more efficacious compounds that can target the molecular heterogeneity of HCC directly addresses an unmet clinical need.”

— Samuel So, The Lui Hac Minh Professor

“We know this disease is hard to tackle. There have been so many failures to identify new drugs in the last 20 years. And I think the AI component helped us to probe the problem differently and more systemically.”

— Xiaopeng Li, Associate professor

What’s next

The researchers have shared their code and developed a web portal for other researchers to use the GPS platform for virtual compound screening. They hope this approach can be applied to discover new therapeutics for a wide range of diseases.

The takeaway

This study demonstrates the power of interdisciplinary collaboration and the use of machine learning to accelerate the drug discovery process. By combining expertise from computer science, biology, and clinical medicine, the researchers were able to identify promising new compounds for treating two difficult diseases, highlighting the potential of this AI-driven approach to drive innovation in therapeutic development.