Shana Kelley says computation can take a more direct route than the random walks of experimental research, but results must go back to the lab for validation. She calls it a virtuous cycle of computation and lab work, not a replacement for the lab when bringing therapies forward.
Shana Kelley says CZI is not aiming for three to six extra months of life for cancer patients. The organization wants new treatment paradigms, early prevention when inflammation first sparks, and a dramatic reduction in disease burden across the whole population instead of end-stage care.
Shana Kelley says the original Biohubs in the Bay Area, Chicago, and New York were reorganized into a single organization called Biohub. Work is now assigned by who has the right expertise, not by geography, so teams can plug the best people into each problem.
Shana Kelley says inflammation is dynamic and must be studied in intact tissue with spatial context, not from single petri-dish cells or snapshots. Biohub Chicago is building implantable sensors that track five to ten inflammatory markers and live-tissue omic platforms to capture the molecular landscape.
Shana Kelley says large language models are useful for protein sequences, but language is linear. Real biology is multidimensional and nonlinear, so Biohub is prepared to build new AI approaches, drawing on its computer science experts, when existing models cannot capture it accurately, as needed.
Shana Kelley says CZI has long shared single-cell databases with the community and will keep pushing data out. The organization always preprints its work so others can see results before publication and use the insights to accelerate their own research instead of keeping them internal.
Shana Kelley says working at the interfaces of chemistry, biology, physics, and engineering lets researchers approach complex problems without decades of baggage. Fresh perspectives from other disciplines produce completely new ways to address old biological problems, a dynamic she calls incredibly powerful and very fertile.
