The Virtual Biology Initiative sparks thoughts on richer biological datasets, more reliable AI models, and what researchers might learn from unexpected results.
A prediction helps us think through a decision. When we claim a model will accelerate discovery, what happens when somebody acts on its output, and what do we learn from the awkward results?
AI depends on physical systems and the people who can build, adapt and maintain them. A companion essay to my AIChE Day at OAU keynote on chemical engineering’s role in Africa’s future.
Before choosing a model, I want to know what it needs to help us decide. Which mutation should we test? Which pathway explanation can we rule out? Where should a sensor go? What should a controller do when the load changes?