On January 20, 2026, the LIGAND-AI consortium officially launched as one of the largest public-private partnerships dedicated to AI-driven drug discovery, backed by more than 60 million euros in funding from the European Union and industry partners through the Innovative Health Initiative (IHI). The five-year project brings together 18 partners across nine countries, with University College London (UCL) serving as a lead academic partner.
Mission: Open Data for AI Drug Design
The consortium's primary objective is to generate large, open, high-quality datasets of how molecules (ligands) bind to proteins, and to use these datasets to train AI models capable of predicting candidate molecules as suitable binders for thousands of human proteins. The emphasis on open data addresses one of the most significant bottlenecks in AI drug discovery: the scarcity of publicly available, high-quality molecular interaction data that can be used to train and validate AI models across the industry.
Leadership and Partners
The consortium is led by Pfizer and the Structural Genomics Consortium (SGC), and includes experts from academia, the life sciences industry, technology companies, and research organizations. The project will investigate thousands of proteins relevant to existing and unmet disease areas, including rare diseases, neurological conditions, and oncology. UCL will contribute expertise in structural biology, computational chemistry, and AI model development.
Addressing the Data Bottleneck
Current AI drug discovery models are often limited by the quantity and quality of training data available. Many pharmaceutical companies hold proprietary datasets of protein-ligand interactions that are not shared with the broader research community, creating fragmentation and redundancy. LIGAND-AI aims to break this cycle by generating standardized, experimentally validated datasets that will be made freely available, enabling AI developers worldwide to build more accurate and generalizable models for drug design.
Technology Approach
The consortium will employ a combination of high-throughput experimental methods, including X-ray crystallography, cryo-electron microscopy, and biophysical assays, to systematically characterize protein-ligand interactions at scale. These experimental data will be paired with computational methods, including molecular dynamics simulations and AI-driven analysis, to create comprehensive binding profiles for each target protein.
Broader Impact
By producing open datasets and validated AI models, LIGAND-AI aims to democratize AI drug discovery, making advanced computational tools accessible to academic researchers, small biotechs, and organizations working on neglected diseases that may lack the resources to generate their own proprietary datasets. The project represents a significant investment by the European Union in maintaining competitiveness in the rapidly evolving AI drug discovery landscape, where U.S. and Chinese companies have attracted the majority of venture capital investment to date.