Professor Hui Peng of the Department of Chemistry and his research group are part of an international consortium led by Structural Genomics Consortium (SGC) who have received more than €60M in funding from the Innovative Health Initiative (IHI). This five-year LIGAND-AI project (2025-2030) aims to systematically identify small molecules that bind to human proteins and generate open, high-quality datasets to advance artificial intelligence (AI)-driven drug discovery.
The LIGAND-AI consortium brings together 18 partners across nine countries to generate open, high-quality datasets and use them to train AI models capable of predicting drug-like molecules as binders for thousands of human proteins.
The project will start by testing up to 0.5 million compounds across approximately 2,000 human proteins. This effort is designed to generate consistent, high-quality binding data that can serve as chemical starting points for downstream drug discovery and as training datasets for machine learning models.
“Right now, most medicines are developed using about 700 human proteins,” Peng explains, “But the human body contains over 20,000. That’s a lot of unexplored potential.”
Peng notes that recent breakthroughs in AI, such as the development of Alphafold2 for protein structure prediction, would not have been possible without decades of high-quality of experimental structural data generated by structural biologists worldwide. “Despite advances in AI-driven protein design recognized by the 2024 Nobel Prize in Chemistry, the development of AI algorithms to predict protein-ligand interactions remains in its infancy.”
The LIGAND-AI project offers similar game-changing potential by homing in on those very proteins. By building and making these standardized datasets ;openly accessible LIGAND-AI will enhance AI-driven drug discovery efforts
Prof. Peng emphasized the multi-national scope of the project. SGC-Toronto, which is the largest experimental site within SGC, will generate testable amounts of targeted proteins. Peng, with his collaborators, will focus on using an automated chemistry method for high-throughput screening of small molecule binders from the designed library.
A key component of this effort is enantioselective protein affinity selection mass spectrometry (E-ASMS), a methodology developed by Xiaoyun Wang, a postdoctoral fellow in the Peng group, and published in 2025: “Enantioselective protein affinity selection mass spectrometry (E-ASMS)." The approach addresses a significant challenge within chemical ligand discovery for ‘undruggable proteins’, by enabling high-throughput and sensitive detection of binding interactions.
Supported by the IHI LIGAND-AI project, the Peng group is currently establishing a high-throughput chemical ligand screening lab in the Lash Miller Building, equipped with high-resolution mass spectrometers and robotic systems, aiming to screen more than 400 proteins per year.
As someone whose lab focuses on studying small molecule protein binding, Peng is excited about this project. “Because it is a very interdisciplinary project, we’re working with many leading experts on structural biology, medicinal chemistry, and AI researchers, from Europe, the U.S.A. and Canada. We have been dreaming of systematically mapping protein-chemical interactions on the genome level. That cannot be achieved by a single team."
In parallel with drug discovery applications, the Peng group will explore the bindings of the same proteins to the whole chemical exposome of humans, including millions of endogenous metabolites and xenobiotics.
With the Peng Lab and other partners from academia and industry contributing this deep expertise, LIGAND-AI is positioned to strengthen AI-enabled drug discovery and expand the accessible landscape of human protein targets.
We have been dreaming of systematically mapping protein-chemical interactions on the genome level. -Prof. Hui Peng
Related Links: