The Virtual Biology Initiative aims to create AI models capable of simulating living cells. This project seeks to enable researchers to conduct digital experiments, potentially reducing the need for extensive physical lab work and high costs.
The initiative has secured a total commitment of $1.8 billion, which includes funding, existing datasets, computing resources, and new measurement technologies. The effort is described as the largest coordinated attempt to generate AI-ready biological data to date.
Key contributors include Google DeepMind, Isomorphic Labs, and Meta, which have collectively committed $300 million. The U.S. Department of Energy plans to invest over $500 million during the next five years, while the National Institutes of Health is contributing datasets valued at more than $500 million.
NVIDIA is providing technical support and computing infrastructure for the project. Other participating research organizations include the Broad Institute, the Allen Institute, the Human Cell Atlas, the Human Protein Atlas, the Gladstone Institutes, and the Wellcome Sanger Institute.
The project will utilize techniques such as spatial transcriptomics, advanced microscopy, and cryo-electron tomography to collect data on cellular responses. Biohub expects the first dataset to be available in approximately one year, with predictive models potentially emerging within five years.
