Scaling Cheap In‑Vivo Causal Testing for Age‑Related Diseases
AI has made hypothesis generation in bio cheap. Anyone can get an answer to ‘could this play a role in my disease’, but how do we go from ‘could’ to ‘does’? The scarce resource now is causal evidence to test hypotheses in vivo. What if that became cheap too? Imagine being able to predict, before running a full medchem campaign, whether a novel therapeutic candidate is likely to causally improve a specific age-related disease. At @unlockscience26 I spoke about this general problem, and the solution we're building at @GordianBio: scaling up pooled in vivo perturbations across organs to build an atlas of causal effects across cardiometabolic disease. Much more to be done, and cool approaches in cancer and other areas from @Ronalfa, Aviv Regev, and others.
Scalable Hypothesis‑testing Datasets Will Drive Biology’s AI ROI
Which AI-enabling datasets will bring the biggest ROI in biology? One type will be "ways to confirm hypotheses at scale". @GordianBio is building atlases of how every expressed gene target affects organs suffering diseases of aging.
Choose the Right Design, Not One‑Size Truth
n=1 is can work for cancer and infectious disease, but often you need a clinical trial. We need to always be asking what the best tool is to get to truth. Don't assume that one experimental design is always correct. And don't...
Analyzing Causal Datasets to Uncover Disease and Aging Mechanisms
Now lets do this for the datasets that contain causal information on how diseases and aging work
AI-Driven Biology Faces Jagged Data Frontier; Initiative Leads
AI accelerating bio is going to be a jagged frontier driven by availability of relevant data. This is a fantastic initiative to get ahead of that. (see also what OAIF @JacobTref is doing).