€60 Million LIGAND-AI Consortium Launches to Accelerate Drug Discovery Through Open AI Datasets
核心洞察
The LIGAND-AI consortium has launched with a €60 million budget to generate large, open datasets of protein-ligand interactions and train AI models for predicting drug candidates across thousands of human proteins (搜索).
Led by Pfizer and the Structural Genomics Consortium (搜索), the five-year project brings together 18 partners across nine countries to target rare neurological, cancer (搜索), and other unmet disease areas.
The initiative aims to transform the expensive and uncertain drug discovery process by combining advanced laboratory technologies with computational methods to create billions of data points for global researchers.
A groundbreaking €60 million artificial intelligence-driven drug discovery initiative has launched, bringing together 18 partners across nine countries to revolutionize how new medicines are developed. The LIGAND-AI consortium, led by Pfizer and the Structural Genomics Consortium (搜索) (SGC), aims to generate massive open datasets of protein-ligand interactions and train advanced AI models to predict candidate molecules for thousands of human proteins (搜索).
The five-year project, funded by the European Union and industry partners through the Innovative Health Initiative (IHI), represents a major shift toward open science in pharmaceutical research. University College London (UCL) serves as the lead academic partner in the UK, working alongside universities in Canada and Germany to spearhead community engagement and expand researcher networks globally.
Addressing Critical Challenges in Drug Development
Early drug discovery currently faces significant obstacles, with scientists spending years testing thousands of molecules to identify just one that binds effectively to a disease-related protein. The process is characterized by long timelines, substantial investment, and considerable uncertainty that LIGAND-AI seeks to address through systematic data generation and AI-powered prediction.
"Machine learning will accelerate the discovery of new medicines. But for that to happen, we need very large, high quality, public datasets so that we can train the algorithms effectively," said Professor Matthew Todd from UCL School of Pharmacy. "This project helps to generate that dataset of how billions of molecules bind human proteins (搜索) - experimentally, in the lab."
The consortium will investigate thousands of proteins relevant to existing and unmet disease areas, including rare neurological conditions, cancer (搜索), and other oncological diseases. By combining advanced laboratory technologies with computational methods, LIGAND-AI aims to create a seamless pipeline from experimental data to predictive modeling.
Open Science Infrastructure and Global Collaboration
The project's commitment to open science sets it apart from traditional pharmaceutical research approaches. All data generated through LIGAND-AI will be shared according to FAIR principles, ensuring information is findable, accessible, interoperable, and reusable by the global scientific community.
"This project brings together scientists and companies from across disciplines within an open science ecosystem," said Professor Aled Edwards, CEO of the Structural Genomics Consortium (搜索) and project coordinator. "It is heartening to see these diverse scientific communities coalesce around a common vision to generate and share valuable chemical data openly with the world."
The consortium includes diverse expertise spanning protein science, structural biology, chemistry, and machine learning. Partners range from academic institutions like the European Molecular Biology Laboratory (搜索) and University Health Network to industry leaders including AstraZeneca, Novo Nordisk, and Thermo Fisher Scientific (搜索).
Industry Partnership and Technological Innovation
Abcam (搜索), a key industry partner, will contribute specialized expertise in recombinant protein engineering, synthesis, and production. The company's capabilities in delivering high-quality, well-characterized proteins at scale are essential for generating the robust ligand-binding datasets required to train predictive models.
"Ligand-AI represents the next frontier in drug discovery—combining cutting-edge AI with large-scale experimental science," said Alejandra Solache Diaz, Senior Vice President of Research & Development at Abcam (搜索). "We're thrilled to bring Abcam's protein capabilities into a consortium of global leaders, working together to build predictive models that will unlock thousands of new targets."
The consortium will generate billions of data points using complementary screening technologies, enabling researchers worldwide to develop, train, and benchmark AI models that predict molecular interactions. This approach promises to dramatically reduce the time and resources required for early-stage drug discovery.
Strategic Vision and Future Impact
LIGAND-AI represents a significant milestone toward the Target 2035 initiative's ambitious goal of discovering chemical modulators for every human protein by 2035. The project will establish shared, open-science infrastructure for AI-driven drug discovery while training a new generation of interdisciplinary scientists fluent in both computational and experimental approaches.
Beyond data generation, the initiative will foster an open discovery ecosystem through collaborative challenges and benchmarking campaigns. The UCL team, led by Professors Matthew Todd and Nicola Burgess-Brown, will spearhead community engagement efforts, seeking donations of protein samples and machine learning models from researchers worldwide.
The project's emphasis on transparency and accessibility aims to ensure that progress is cumulative and benefits the entire scientific community. By reducing fragmentation across sectors and catalyzing global collaboration, LIGAND-AI could fundamentally transform how new therapeutics are discovered and developed for patients with unmet medical needs.