Computers gamers solve protein puzzle
Online computer gaming is sometimes viewed as a mere pastime without much outward benefit, but a new paper published today in the journal Nature Structural and Molecular Biology reveals how gamers playing an online game called Foldit have managed to crack the 3-dimensional structure of an important protein produced by the Mason-Pfizer Monkey virus, which causes a disease similar to AIDS in monkeys.
Figuring out the exact 3-D structures of proteins is really important in order to understand how they work, and also to develop drugs that can target them effectively - regular listeners may remember the paper's lead author Professor David Baker talking about this on the show back in August. There are several lab techniques that can be used to figure out protein structures, but these don't always provide a definite answer, and tend to rely on having a good model with which to intepret the physical data. In the case of the monkey virus protein, scientists had struggled for a long time to come up with a solution with no luck. So the researchers turned to the ingenuity of the Foldit players to try and come up with the 3-D structure.
There is a lot of computing power going into uncovering the 3-D structure of proteins - there's a big server called Rosetta, which also uses distributed computing power - that's the power of people's home computers when they're not in use - to churn through millions of possible protein structures in an automated way to look for the ones that seem most likely. But again, in some cases - such as this monkey virus protein - Rosetta still isn't providing the right answers, and a little bit of human intuition and puzzle-solving is needed.
In this case, the researchers gave the Foldit gamers some basic information about the protein's likely structure, based on data from a technique called NMR spectroscopy. The gamers set to work, and came up with a model, which they then tweaked. The researchers particularly note the contributions of three players - called spvincent, grabhorn and mimi - for making specific breakthroughs in solving the structure.
Once the players had come up with a good model, based on sound biochemical and physical principles, the researchers could then use that model as the basis for interpreting the data from physical analysis of the protein, using a technique called X-ray crystallography. They found that it was a really good match, proving that the Foldit gamers had accurately predicted the structure of the protein.
In the case of the monkey virus protein, the final structure has revealed some interesting regions that could be targeted by specifically-designed drugs. But more importantly, this is the first demonstration of the power and ingenuity of online gamers to solve long-standing scientific problems by combining computing power with the human brain. And this doesn't necessarily need the brains of trained scientists, as most of the Foldit gamers don't have a background in biochemistry. And given that there are many more unsolved protein structures out there, it's likely that the Foldit gamers are going to make a lot more breakthroughs in the future - and it's nice to know that they're making a big contribution to science while sitting at home on the computer.