As you shall soon see, this question might have gone under the general questions section. But I wasn’t sure, so I put it here. Anyways, I certainly am concerned about the destruction of the rain forest. And I certainly think we should do all we can to save it while we can (just to show you my sincerity when I say this, I go to a free click on site every day to save a parcel of rain forest). And as many you probably know, ever since I was a child, I have liked to think up devices that have various uses. One such device I have thought of is at least a partial solution to the rain forest dilemma. Allow me to explain.
The more we lose of the rain forest, the more we lose the beneficial drugs and medicines that the plants and animals in the rain forest contains. So my “device” to deal with this loss would actually be a computer program that generates random molecule configurations and then tests them to see if they have any use. Of course the program would have to be adapted so that it wouldn’t just be chugging out totally random molecules, but molecules that you’d expect to find in biology (or in a rain forest, as I’ve said).
The only problem I see with my program idea is that it almost seems too easy. I can’t help but think that someone should’ve thought of it by now if it is really a viable idea. So my question is, could something like my idea work? Why hasn’t it been tried? And if it has been tried, who tried it? And how successful was it?
Intriguing, but I doubt it could work. AFAIK the “behaviour” of large organic molecules is not sufficiently understood to enable simulation. Even supposing that the programming task were easy (which I doubt) I rather suspect the sheer number of combinations would defeat the number-crunching capablities of computers for the foreseeable future.
IANABiochemist, but I’m guessing that something complex in the wild with potential to be used as a medicine would have a useful effect when:
it’s a particular isomer of the molecule
in a specific concentration
in the presence of a number of other complex catalysts also at specific concentrations
in the absence of certain other chemicals
over a narrow temperature range
over a narrow pressure range (etc)
It’s not really possible to simulate the interactions of large molecules to that degree; things can be done with computers, but keep in mind that drugs can act on any of millions of chemical pathways in the body, and simulating the chemical workings of an entire human being would be impossible.
Besides, it wouldn’t help much if we couldn’t produce the drug. A lot of plausible chemicals would be difficult or impossible to produce without a natural source or precursor. Generating molecules from scratch is not very realistic.
There’s also the computer time. It costs a lot of time and money to do much of anything with large molecules–we’re talking supercomputers or at the very least a Beowulf cluster. By the time you’re done running models on some of these macromolecules you would’ve been better off trying to come up with a way to design it and test it in lab.
You might want to check out Folding@Home, which is a distrubuted client program that models protein folding, which is very import for medical application.
