How accurate is folding@home to the real thing?

[castlebomb44]

How accurate is folding@home to the real thing? And how do you check if the folding is what is going on in a test tube somewhere?

[bruce]

How accurate is folding@home to the real thing? And how do you check if the folding is what is going on in a test tube somewhere?

If you're scientifically inclined, read the papers that have been published so far. http://folding.stanford.edu/English/Papers

The purpose of FAH is to better understand the mechanics of the folding process. Results are generally compared to laboratory results and if they do not agree, somebody has to come up with a better model of what's actually happening -- forcing the scientists to understand the process better.

You might be familiar with the history of astronomy. Long ago there was a theory that the earth was the center of the universe and there were complex explanations for the apparent motions of the planets. Newton devised some mathematical laws and Galileo made some observations and proposed that the Sun was the center of the universe. After many years (and a couple of excommunications) that became the accepted explanation. Some new observations changed the theory again placing the sun in a fairly remote part of the galaxy. More observations and some new Mathematics by a guy named Einstein and the model of the universe changed again.

Such is the progress of developing better and better models which provide better and better explanations. At each stage of development, there was a model that was "accurate" enough to explain the observations that had been made and there was a constant back-and-forth interchange between observations and mathematics.

Protein folding has gone and will probably continue to go through a similar set of steps.

[uncle_fungus]

Take a look at paper 8 in particular, which directly addresses the question.

Protein folding is difficult to simulate with classical molecular dynamics. Secondary structure motifs such as a-helices and b-hairpins can form in 0.1–10 ms, whereas small proteins have been shown to fold completely in tens of microseconds2. The longest folding simulation to date is a single 1-ms simulation of the villin headpiece; however, such single runs may miss many features of the folding process as it is a heterogeneous reaction involving an ensemble of transition states. Here, we have used a distributed computing implementation to produce tens of thousands of 5–20-ns trajectories (700 ms) to simulate mutants of the designed mini-protein BBA5. The fast relaxation dynamics these predict were compared with the results of laser temperature-jump experiments. Our computational predictions are in excellent agreement with the experimentally determined mean folding times and equilibrium constants. The rapid folding of BBA5 is due to the swift formation of secondary structure. The convergence of experimentally and computationally accessible timescales will allow the comparison of absolute quantities characterizing in vitro and in silico (computed) protein folding.

[jheil]

That is cool, thanks for posting the excerpt I didn't feel like reading through the paper plus the stuff I did read was a little over my head on the biochemistry and kinetics. Given the comparison between in silico and experimentally determined sturcture we can see that the in silico structure determination is quiet good given the incredible difficulty of modeling such a system. However different proteins might not work out so well. I understand that folding at home cores all are using a priori modeling. I was wondering if homology modeling might be useful for gaining confidence in structures that aren't experimentally determined but have the 40-50% identity needed for homology modeling?

Also, experimentally determined proteins are usually crystallized so how do we even know if we are actually getting biologically active structures? In Silico modeling might even be good for confirming if known structures are even the biologically active structures.