It's just about every way you could imagine. I work at a large company, where my department has something like 75 people in 25 labs working on one disease area.

It's just about every way you could imagine. I work at a large company, where my department has something like 75 people in 25 labs working on one disease area. Anything we think we can build a case for is worth considering, and we'll run down lots of leads that don't turn up anything. We get chemists involved to make tool compounds that might hit your target pathway, and eventually, if you build enough evidence to convince management, you'll put together a team. That team will involve a few biology labs, a chemist (could be a biochemist) or two, a toxicologist, and a pharmacokinetics person (ADME). Then, if it's a chemical target, they'll make thousands of compounds to test in their preliminary tests to make sure the compound very basically attaches to the target. The ones that make it past that might make it into a cell-based assay then rodents. The path each treatment takes through specific assays is different for every project depending on the specifics, but it generally follows that progression.