Slashdot Mirror
Folding@Home 2.0 - An Online Protein Folding Game
a boy named woo writes "Tired of justifying your gaming addiction? Now you can really help accomplish something while you play... thanks to Howard Hughes Medical Institute researcher David Baker at the University of Washington." In collaboration with others, Baker has designed a game, called "Foldit," with a practical outcome: players manipulate on-screen images of protein chains and attempt to predict their folding patterns. From the article:
"'Our main goal was to make sure that anyone could do it, even if they didn't know what biochemistry or protein folding was,' says [co-creator Zoran] Popovic. At the moment, the game only uses proteins whose three-dimensional structures have been solved by researchers. But, says Popovic, 'soon we'll be introducing puzzles for which we don't know the solution.'"
No, it's the other way around. There are many ways to fold it so folding is easy. But there is only one solution with the lowest (free) energy. The number of ways to fold is very large. To determine if your solution is the lowest, you have to check all possible ways of folding. So in this game, they'll let you fold and if you are better than all the human and computer opponents for a certain period, you probably get some points.
DNA is the ultimate spaghetti code.
Could be the next test of the true talent of a programmer. These have to be the most complex 3D problems ever. If a person were good at it , would probably be a paying occupation.
Folding@Home on my PS3 take a couple hours at least per nanosecond of folding. A work unit is not a fold. It's a tiny fraction of a fold which can take thousands of nanoseconds. If a human can solve it in a day, that's a VAST improvement.
Well if the company is smart they'd give him a job or scholarship or such. Anyway afaik drug companies never find proteins, they always "steal" them, i.e. patent the work of government paid university researchers, or buy the patents from the university for peanuts. All the drug companies actually do is the FDA "paperwork", which is actually quite costly, and the marketing.
In an ideal world, if the fed. gov. paid for the universities to do the research, they could also pay for the universities to get the drug some preliminary FDA approval. After that, any U.S. based (generic) drug company could produce the drug (completing their part of the approval process). However, only U.S. based companies would have this right.
The Christian religion has been and still is the principal enemy of moral progress in the world. -- Bertrand Russell
Do post back here, then, when the Linux version is ready, please...
+1 Funny? Or just another fucking idiot? It's so hard to tell these days...
Have you tried playing the game? You're pulling atoms around, but you also have the same toolbox that the computer has when it's trying to fold a protein, including gradient-based minimization, side-chain repacking, and cavity detection. The idea behind fold.it seems to be that humans might have better judgement about how to applies the tools used for protein structure prediction.
Also, Folding@Home takes an extremely long time to fold proteins of any reasonable size compared to a program like Fold.It. The submission title is a misnomer: Folding@Home is totally unrelated to Fold.It, they're totally independent pieces of software that represent very different approaches to molecular simulation.
Seriously: I know that's a bit subjective, but I (and literally millions of others) disagree with that. I run BOINC on my computers because it does *not* suck. I run several projects on each, ensuring that when/if one has a server outage or workunit lull, my BOINC clients are doing something useful for someone *else*.
Perhaps you've never heard of or used the account managers BAM or GridRepublic; they make managing projects and client computers a breeze. They're easy to use but quite flexible.
I wholeheartedly think their functionality should have been built into BOINC from the get-go, but perhaps the authors of BOINC (the Berkeley team that does seti@home) omitted it intentionally because *someone* has to run the software and they didn't want to be the ones.
Folding@home is a big and well-funded project that doesn't benefit from the middleware that is BOINC, but they also reap no benefits from the network effect that such a wide userbase offers.
So no, I don't think BOINC sucks anything like "really really bad". If you're happy with Folding@home, fine; but don't slander BOINC and thus the dozens of projects that use it simply to re-enforce your own decision, especially in such a shallow way. (TWO "really"s with a "suck", and no other commentary? This guy must be both serious and knowledgeable.) BOINC ain't perfect, but it's pretty good already, and continues to improve over time.
The "game" has two options to automatically "shake and wiggle" a molecule for collisions and misalignments a computer can easily identify. You don't have to handle trivial collisions by hand.
But there are certain problems that are easy for a person because humans can visualize and imagine a structure, something a computer simply cannot. This is exactly what this program is about. You look at such a molecule and can easily determine that bending it here or there allows you to crunch it further. A computer would have to try all, or at least many, combinations that you already exclude as pointless just from looking at them.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
If by paperwork, you mean testing procedures that cost hundreds of millions of dollars and often result in duds, then you're spot on.