ChemHopeful said:
I'm always hearing about monolayers... what make them so interesting?
Kind of depends on what kind of monolayer you're talking about. If you're talking about a monolayer of molecules adsorbed to a surface (a Langmuir monolayer) then it's an interesting thing because it's a relatively simple and well-understood situation. So if you want to study surface films, then your starting model is usually going to be a monolayer.
And while I'm asking dumb questions, why are 2 out of any 3 physical chemists working on protein folding modeling?
Well if everyone who works on protein folding is a p-chemist, then I guess that's true. I wouldn't really count most of them as physical chemists though. I've got some of 'em working further down the hall, and a number of them aren't chemists of any kind, but rather computer science guys who've been brought up to speed on the basic chemistry and physics.
Anyway, protein folding is important because protein structure is important. The function of proteins is pretty much
all about structure. I mean, that's literally the difference between scrambled eggs and something that can develop into a chicken!
The experimental methods of getting the 3d structure of proteins aren't that great. The most widely used being X-ray crystallography. Which means that first, you have to produce a lot of your protein and purify it, which is a lot of work. Then you have to crystallize that into nice well-ordered crystals. Which is really a dark art, and can involve playing around with all kinds of solvents, or perhaps binding the protein to an antibody.
Then you have to hope and pray that that procedure doesn't screw up or warp the protein too much, and that the crystallization itself doesn't warp the protein too much. Then you blast the thing with X-rays, which causes ionization and radical formation and can reduce metal ions. And so you hope that the radiation damage is minimal. Finally you get out a diffraction pattern which computation renders into an electron-density map, and then you use a model (not unlike those used with protein folding) to fit the structure to that map as best you can. And the final result is, if you're lucky, a structure where the atoms are off by an average of half an angstrom or so :)
Worse, a substantial portion of proteins are membrane proteins, which means they sit in a membrane, and thus have a hydrophobic middle. Take them out of the membrane to try to crystallize them, and they have a strong tendency to try to turn themselves inside-out to get that hydrophobic bit away from the solution!
So it's difficult and at best, time-and-labor consuming. The other main method is NMR, which has the benefit of being able to study the things in solution, but its own set of drawbacks and headaches (for one, you want your molecules to tumble quickly, which proteins don't).
There are more exotic methods, like neutron diffraction. But http://en.wikipedia.org/wiki/Category:Neutron_facilities" aren't exactly easy to come by.
In the human body alone, there's over a million different proteins, and we'd like to know the structure of all of them. And computer modeling provides the only automated way of doing it.
This, BTW, is representative of what some people are calling a new branch of science. Formerly, you had 'theory' and 'experiment', but some think we should now add 'computation' to that. Because protein folding (and quite a few other things) don't quite fit into 'theory'. After all, the protein-folding folks aren't really working on the theory of protein folding. The underlying physics is completely known. They're doing the calculations and improving the ways to do the calculations.
