answer:Slowly. I have a cousin who is in cancer research on a microbiology level. In general they start out with a hypothesis – with Condition X, knowing how organisms usually react, trying Compound Y and Compound Z ought to have the following effect. Then they set up a couple of hundred samples, apply the compounds to the organisms, wait (weeks, maybe months) and see if their hypotheses are correct. If not, they adjust amounts, levels, temperature – various variables, and try again. To me, it’s painstakingly frustrating, but ultimately it’s how science is done – you try something, see why it failed, try something else, and so on.