NEW YORK – The Chan Zuckerberg Initiative said Thursday that it has launched the Billion Cells Project, a collaboration with 10x Genomics and Ultima Genomics to create a dataset to train AI models in biology.
"Biology not only needs more data, the field needs more data faster and in interoperable formats to support AI models that address specific problems," Jonah Cool, senior science program officer for cell science at CZI, said in a statement. "This project represents a unique approach to scaling and standardizing scientific outputs for AI and more."
The project is also partnering with academic researchers, including Alexander Marson, director of the Gladstone-UCSF Institute of Genomic Immunology. The collaborators will use 10x Genomics' Chromium GEM-X technology for single-cell analysis and Ultima's UG 100 for sequencing.
Financial and other terms of the collaboration were not disclosed. CZI said in a statement that it anticipates additional partnerships with industry and academia "to ensure delivery of the ambition of the project on a short timescale."
Advancing computational tools for biologists has been a longstanding goal at CZI, and the organization has already created tools to help researchers access new, generative AI virtual cell models that have come out in the last couple of years.
Now, data generated from the Billion Cells Project will be used to train new virtual cell models using CZI's computing expertise. CZI said it plans to make the results "open source and freely available to help scientists around the world make new discoveries about human biology." Collaborators will generate data across multiple domains, beginning with mouse, zebrafish, and human primary cell models.
"This project will provide a necessary scale of data to understand the functional effects of human genetic variants and characterize the genetic drivers of human disease," Marson said. "Ultimately, the Billion Cells Project will also be a functional roadmap to guide drug development, identifying targets to restore diseased cells to health."
Michael Schnall-Levin, 10x Genomics' cofounder and chief technology officer, said in a statement that the firm would provide "enhanced support, collaboration, and access to our internal experts in R&D, bioinformatics, and data analysis."
"We are working closely with our partners CZI and 10x to ensure a first-rate customer experience for this project," Ultima CEO and Cofounder Gilad Almogy said in an email, including providing "improved analysis." Late last year, Ultima Genomics provided sequencing services for Vevo Therapeutics and Parse Biosciences as they generated a 100-million-cell atlas for use in AI models for drug discovery.