Biohub announces $1.8 billion collaboration for virtual cell data

Biohub has announced a $1.8 billion collaboration with the Department of Energy, National Institutes of Health, Meta, and Google to collect open-source, multi-scale biological data for developing next-generation virtual cells. This initiative is heavily coupled with the Human Cell Atlas, an international consortium focused on dataset generation. Fabian Theis, who leads AI and computational biology efforts at Helmholtz Munich, noted that the field is shifting from mapping cell atlases to prediction and now to steering experiments.

Theis highlighted that data generation is the primary driver for renewed pharma interest in virtual cells, citing the explosion in perturbation data scale, such as 10 million transcriptomes in single experiments and full CRISPR screens across 100 different cell lines. He argued that while simple linear models often suffice for basic cell type prediction, complex nonlinear models are required for harder tasks like predicting the effect of perturbations or spatial neighbor gene expression. "Biohub just announced a $1.8 billion collaboration with the DOE, NIH, Meta, and Google to collect an open source, multi-scale biological data that can be used to develop a new generation of virtual cells," Theis said.