I haven't worked on the drug-side of things, but here my bio perspective: It's kind of out-of-vogue, but consider the "lock and key" model of proteins and small molecules (drugs). For drug design, what you want to do is get a key that fits just one lock (to pull whatever lever) and not others (to avoid side-effects). It's relatively easy to find a molecule that fits a protein, because that protein is what you might spend years researching and probing, but it's tricky to check if it does anything against ~100,000 others in humans. If you could do an in silico computational survey to be like, oh, maybe it'll target this accidentally, you could spot-check those in vitro, and/or stick on some other atoms to your small-molecule to make it not fit that off-target.

Holy grail, IMO, though is being able to design de novo protein sequences (to make "biologics", aka engineered protein drugs) that can a) target (bind/block/enhance) or do (chemical reactions) what you want and only that, b) are easily synthesizeable by bacteria/yeast (cheap to make), and c) are stable (easy to transport/store).