Pondering further on the subject of genetic engineering's progress into the realm of plausible activities for a small, even hobby-level, lab, an idea I had back in the 1980s starts sounding close to reality:
How practical would it be to splice hemp genes into oregano?
It doesn't have to be hemp and oregano, of course. It could be white poppies and California poppies, or magic mushrooms and market mushrooms, or even coca and cola. Just so you get a plant which looks just like something that's commonly and legitimately cultivated, and which incidentally produces an illicit drug.
It seems that such an enterprise would be profitable enough to justify the investment, assuming that it could actually be done.
Probably not on the drug lords' radar is the notion of splicing dope-producing genes into an otherwise innocuous strain of E. coli - "here, eat this, and it'll keep you hopped up for the rest of your life." (On the other hand, bacterial dope production, using bacteria not suited to such resident production, would have advantages; bacterial cultures are a lot easier to hide than poppy fields. You just have to watch out for nasty suspicious people who might think you were using those supplies to grow anthrax.)