Google DeepMind partners with Chan Zuckerberg Virtual Biology Initiative to build biological datasets

Pushmeet Kohli announced that Google DeepMind is partnering with the Chan Zuckerberg Virtual Biology Initiative to produce the multimodal datasets required for improved predictive biology models, addressing the need for experimental biological data at scale.

Kohli noted that advancing predictive biology requires understanding life across multiple scales, from DNA and individual proteins to living cells and whole-body health. Citing precedents such as AlphaFold and the Protein Data Bank, the effort relies on open, high-quality data and aligns with work by the CZ Biohub initiative, as detailed in the Virtual Biology Initiative announcement.