Folding@Home Reports Success 344
msheppard writes "This Article describes how the folding@home distributed computing project is reporting that they used the data processed on client machines to "predict the folding rate and trajectory of the average molecule." Too bad Seti@Home hasn't had a hit yet."
Oh, *that* kind of folding. (Score:4, Funny)
I fold at home (Score:3, Funny)
Re:I fold at home (Score:4, Funny)
Re:I fold at home (Score:3, Interesting)
Folding @ Home page is (Score:4, Informative)
Links of course (Score:5, Informative)
Folding@Home Home [stanford.edu]
For the real info though check out the Forums [folding-community.org]
Token link to how my team is doing [stanford.edu].
PRIME1
Re:Links of course (Score:2, Informative)
Protein!!!! (Score:5, Funny)
That's average protein molecule, you insensitive clod!
Re:Protein!!!! (Score:3, Funny)
Average molecules come in plain white bags with black stencil lettering "Molecules"
They're located in the science section of your local grocery store.
Re:Protein!!!! (Score:4, Funny)
Re:Protein!!!! (Score:4, Funny)
Its just a shame (Score:4, Interesting)
Re:Its just a shame (Score:3, Insightful)
Try this instead (Score:4, Funny)
go here [distributedfolding.org] instead
Re:Its just a shame (Score:4, Interesting)
Re:Its just a shame (Score:3, Interesting)
I'm not sure how predictive the Nussinov work was considering the methodology.
Finally, I agree that we need to mine the trajectories more. We're doing that, but perhaps more importantly, we will release the data for others to mine too!
V
Other uses for Distributed Computing (Score:5, Funny)
Re:Other uses for Distributed Computing (Score:4, Interesting)
Re:Other uses for Distributed Computing (Score:5, Funny)
--A2K
Re:Other uses for Distributed Computing (Score:2)
well... (Score:5, Funny)
Well, here's the thing: "we know molecules exist..."
Re:well... (Score:5, Funny)
Re:well... (Score:2, Insightful)
dfold too! (Score:5, Informative)
Re:dfold too! (Score:4, Funny)
They could work on the screensaver version (Score:5, Insightful)
SETI@home has much nicer graphics, albeit, a much dumber purpose. I'll stick with folding@home, but I wish they would pretty the damn thing up a little--at least on the Mac OS X platform.
Re:They could work on the screensaver version (Score:2)
Plus I'm not too anxious to let the static images sit on my LCD screen overnight.
Just turn off the damn screen if nobody will watch it for the night. No need to keep it powered on (even if it consumes less than a CRT). It'll probably live longer, and your monthly electricity bill will be a bit lower (~ $2).
But that begs the question: is it possible to turn off the LCD of the new Macs? Or the CRT of the older iMacs? (not meant as flamebait, just asking because I don't have access to one)
Re:They could work on the screensaver version (Score:3, Interesting)
So many to choose from! (Score:4, Interesting)
Re:So many to choose from! (Score:5, Interesting)
Bottomquark has reviewed [bottomquark.com] a number of projects.
And here [distributedprojects.com] is a community site of people participating in such projects.
Re:So many to choose from! (Score:2, Interesting)
Now this [draves.org] is the new official site, and the only place I can find that has a source download.
Re:So many to choose from! (Score:3, Insightful)
I'm really suprised that none of the special-effects CG companies are using this. This kind of grid computing is great for rendering. I could see Pixar doing a really slick screen-saver & maybe letting you see the frames you rendered as a "reward" (maybe not all of them - don't want to give away too much of the movie to a geek with a super powerful computer). It would get their rendering done for free and would be a great promo for the movie. Who wouldn't go to see a movie they helped produce? More than once - " Here comes the frame I rendered... There! Did you see it? Just when Nemo swims up to that shark?... I did that!"
Re:So many to choose from! (Score:3, Interesting)
What bothered me is they weren't responsible for damage to your computer, your data, they wouldn't guarantee the non-existence of virii, and they wouldn't tell you what kinds of things were being computed on your computer (it was processing power as a service - you'd get paid a measly amount to let them use your computer).
How was I to know they wouldn't be using MY computer to do things I'd consider unethical or illegal?
Whatever service you join, be SURE to read the fine print carefully beforehand.
Re:So many to choose from! (Score:3, Interesting)
API???? (Score:2, Interesting)
Article is incorrect... (Score:5, Interesting)
Re:Article is incorrect... (Score:2)
Good CS, bad chemistry (Score:5, Interesting)
Don't get me wrong, the geek half of me thinks that what they're doing is very cool (and far more interesting/useful than Seti@Home). But I don't think it's very relevant to biology, and I doubt it'll ever replace traditional methods. Computers have almost unlimited potential as an aid to experimental structural biology, but in silico protein folding is still a pipe dream and a hand-waving exercise. The theory is really cool, the practical applications are nearly zero.
(Disclaimer: I don't have a PhD so I'm not very qualified in this field, but I do have a BS in biology and a fair amount of experience in programming and some knowledge of molecular simulation.)
Good CS, good chemistry (Score:5, Informative)
First, you can't expect to go from no success to complete success overnight. People have been trying to fold proteins for some time now and have basically failed because it is freakin' hard. The theory is in principle in place, a least to a first approximation, but the calculations are so intensive that they have basically beaten every comer. As an undergraduate I remember how everyone in the field thought getting bigger and better grants and buying bigger and bigger computers was the answer. Oh to be SGI in those days. They sum up the problem pretty well in the Nature paper, essentially a modern (desktop) computer would require a few decades to crunch through a single useful length simulation. Then you need to do it many times to get a useful answer (say 100-1000). Even supercomputers are going to balk at that kind of calculation. Moore's law what it is, we should then be able to get through an in silico simulation in a week on a single computer (when its this fast crystallography really will be dead) by, oh say 2040 at best. (somebody want to calculate that exactly, 10000yrs -> 0.02yrs is how many doubles). So yes, this hasn't gotten rid of x-ray crystallography just yet.
But this is still really cool. Complaints about interface and maintenance aside, this was a great system. It relied on four pretty bright insights.
First, that distributed computing is essentially the poor man's (cough, the academic's) super computer. Also, it automatically adapts itself to technological improvements. People will buy new computers from time to time and, hopefully, reload your software.
Second, that there was no reason other than no one had sufficiently brute forced the process that the existing methods shouldn't work. They use a bunch of 'cheating' techniques to make this managable during the screen saver timescale, such as a united atom model (I think that means they ignore aliphatic hydrogens) and implicit solvent (don't treat it as individual solvent molecules, just a uniform field). It was an open question as to whether this approach would work at all or if you had to go over to much more explicit methods to get it to work at all. It appears that this has kinda worked with the cheater methods in place.
Third, choice of a test case. Yes they chose something that was small. This isn't surprising. They wanted to be done sometime this decade, remember there is a graduate student as the primary author here. Small was necessary. However they also chose a FAST-FOLDING protein. That was clever. Basically, even with distributed computing, it is still hard to simulate a full microsecond of time. Thus they chose something that had some chance of completing its folding one the time scale that they could look at.
Fourth, they remembered their P-Chem. It is really hard to run these things to completion... so they didn't. You don't have to run the simulation until 99% of the molecules have completely folded, just until an appreciable number have folded and you can extrapolate the behavior from that. They ran a 20ns simulation (at the longest). The thing takes 7us for ~60% to fold. As a result only once in a great ong while did one of the simulations actually produce a folded protein. But by doing it ~10000 times they could figure out how that translated into the rate constant. That's clever.
That said, yes there is a long way to go on this, but its still a really clever paper. No we haven't cured cure cancer yet, but its still progress. And forget an in silico structure of the ATPase, that's largely understood already (check the RSCB/PDB [rcsb.org] there's a bunch). The real challenge will be getting a structure that size that hasn't been solved by other methods and convincing anyone else that you're right! Disclosure- I don't have PhD in this area yet, but I'm close.
Re:Good CS, bad chemistry (Score:4, Informative)
(Ignoring for the moment that crystallography has it's own issues. . . at least it can show active sites and quaternary structure)
Well, for the first, we can't throw out crystallography even then. When you're doing a computer calculation, you are in the realm of theory. (even if you have arbitrary accuracy).
You will still need to do experimental verifications now and then.
At the moment, about 2/3 of known protein structures [pdb.org] have been mapped through X-ray crystallography. At best the resolutions are about 1.8 Å, which is pretty good. So you can see quite a bit more than quaternary structure!
The other third is done with NMR spectroscopy,
usually with some powerful computing help to figure it out.
And then there are a pitiful few,
done with computers and experimental data.
These structures also have the poorest accuracy.
Note that computers will never, ever be able to figure out a protein structre ab initio. (i.e. without any info except the sequence)
Do the math, say you have 100 amino acids, and you
test say, 4 conformations for each, that's 4^100
combinations to test.. and you test 10 million a second, it'll take you 5E45 years.
Much older than the current universe.
(Disclaimer: I do not -yet- have my PhD in computational biochemistry.. but I'm working on it..)
Re:Good CS, bad chemistry (Score:3, Informative)
The nifty thing with the folding@home study is that there were able to basically show that invoking a simple physical force field system was enough to get pathways, though they don't make too big a deal about this, maybe someone else has already done this, but I'd be surprised if they managed to do as many trajectories as were done here. I imagine it'll be a while before they process the trajectories to try and find actual pathways (very compute intensive), but the fact that they found a comparable rate (we're not talking global conformation here, these are kinetics) suggests that they may be sitting on top of an actual description of the folding pathway for this teeny protein. Spiffy!
Re:Good CS, bad chemistry (Score:3, Insightful)
Re:Good CS, bad chemistry (Score:3, Interesting)
Besides, it tells you nothing about enzymatic activity, interaction with other proteins, interaction with small molecules and ions. . . these are not easily simulated, and certainly aren't covered by their current method. The software they're actually using is fairly standard (I don't mean this in a bad way), and isn't really the type of thing that would be useful for, say, docking experiments.
Frankly, I find David Baker's work at the U of Wasington to be far more impressive from the perspective of biology. (But still not good enough to replace Xray and NMR methods anytime before I retire.) He's not doing distributed computing though, so I guess it doesn't rise to Slashdot levels of sexiness.
Re:You've got a point.......... (Score:2)
And, to give Pande some credit, I read the Nature paper and I thought it was interesting and a good proof-of-concept. I stand by my intial assessment of the long-term prospects of this kind of research, however.
good place to start (Score:3, Informative)
Re:good place to start (Score:2)
Why do people download these blah@home clients? (Score:3, Interesting)
Then this phenomina should really be looked at by marketing people. It's amazing they can start a project, and just assume people will want to download their little client to use up cpu power ala screen saver.
What drives you people to use these clients? Why bother? And don't tell me it's cuz you want to do your part to find aliens
--Me
People hate to see computers sit idle (Score:5, Insightful)
They have paid for the hardware, paid for the bandwidth, paid for the electricity. It should be doing SOMETHING. Even if it is just displaying flying toasters.
Re:People hate to see computers sit idle (Score:2, Interesting)
Plus, again very interesting for the geeky, even with a (lets be frank) stupid project like SETI, there is still the WHAT-IF factor.
It may be an infinitesimal chance, but just imagine YOU were "the one". Rather alluring.
Re:People hate to see computers sit idle (Score:2)
Of course, as far as rc5 was concerned, I was strictly in it for the cash
Seti? (Score:3, Interesting)
In the "Space" documentary series (hosted by Sam Neill), one researcher mentioned something about "except that one time".
Apparently they had some signal, but it was gone before they could reallign the dishes to get a confirmation.
I guess they ruled out possible "domestic" signals...
If anyone could guide me to a more elaborate source except that remark of one of the researchers, I would like to read it
Comment removed (Score:5, Interesting)
Re:Seti? (Score:3, Funny)
Assuming the Wow! signal is a typical SETI-like transmission, then we can expect valid SETI hits to be very strong, high intermittent signals which appear once (as the transit beam sweeps past Earth), and never repeat again.
So, in other words, they admit that they can never fulfill their own requirements (namely, the repeatability requirement).
Rather daunting, isn't it...
Wow! (was Re:Seti?) (Score:2)
(I presume that this is what they were referring to).
NP-Completeness (Score:5, Informative)
This [berkeley.edu](postscript) is the the original paper on the hardness of String Folding problems.
Make money with folding@home! (Score:4, Funny)
This article was presented to you because you subscribe to the opt-in Slashdot site. If you no longer want to subscribe, please write "unsubscribe" on a $5 bill and send to Envelopes@Home. All removal requests honored.
This article can not be considered spam because we provide an opt-out method per Bill S. 1618 TITLE III passed by the 105th U.S. Congress.
This message is NOT intended for residents of WA, CA, or VA.
Margin of error... (Score:2, Interesting)
Re:Margin of error... (Score:2, Funny)
Re:Margin of error... (Score:4, Informative)
Google Toolbar is now distributed. (Score:3, Informative)
Later, when folding@home has folded, the distributed power of the toolbar may be used to make a 'Super-Google' of sorts. (is that a pun?)
Re:Google Toolbar is now distributed. (Score:2)
Google supporting Folding@Home (Score:5, Informative)
Good results (Score:5, Interesting)
Most current radios, based on decades old tech, broadcast a very orderly signal. It is confined to a narrow band range, only one transmitter is allowed per channel, the data being transmitted is uncompressed and so has many repeating orderly patterns.
To increase capacity future radios will do the opposite.. They will broadcast compressed data that seems completely random, they will use a large swath of spectrum, they will repeat parts of the same signal across a large portion of the spectrum using a "chipping" algorithm. Even farther in the future, so many radios at once may be talking on the same spectrum that to identify a particular sender in order to communicate you'll have to use multiple antennaes and know his location (you'll share spectrum based on location).
What is the end result of advanced communication gear that intelligent minds develop? What is the optimal result? To an outside observer the signal will seem like pure, almost totally random noise. Only to the electronics of a particular receiver that has the correct encryption and chipping key will it seem like anything else.
THAT's why we can't hear anything. Trillions of sentient beings could communicate using this method and we wouldn't hear a thing.
Re:Good results (Score:2)
Interesting idea... And it could be quite truthful if a general path of progress forced civilizations to choose methods of communication with signals nearing background noises or pure white noise...
Maybe it is would be a good idea to change the project into "Hack@SETI". However this could be dangerous. Imagine an orange being carrying two antennas, glasses, dressed on a Yuppie-like fashion and crying: I'm Bell Gadzes from Migrozoft Conglomeration, Red Moon, Vaxinton System, United Empires!.. You violated our DTCA (Digital Trillion Copyright Act)!! You will be assimilated!!!
Re:Good results (Score:2)
To date, only a small fraction of the sky has been searched, this development was going to increase that tremendously.
Re:Good results (Score:3, Insightful)
Also see this item [berkeley.edu] in the Seti@Home FAQ.
Re:Good results (Score:3, Insightful)
Re:Good results (Score:3, Insightful)
A lot of focus is made on SETI and listening for signals. We have in fact in the past sent signals to outer space with a specific message. In 1974, Frank Drake used the radio telescope at Arecibo, Puerto Rico to beam an elaborate coded message in the direction of globular star cluster M13. The message, coded in the binary notation of ones and zeros, contained 1679 bits of information. 1679 is the product of two prime numbers 23 and 73, which should suggest to an alien to break the message up into some combination of those two numbers. When the message is arranged in 23 columns of 73 bits each, and the zeros and ones are replaced by white squares and black squares. Coded into this pattern are from top to bottom: binary representations of the numbers 1- 10, atomic numbers of the five elements essential to terrestrial life, the chemical formula of the DNA molecule, numbers for the average human height and the world's human population, images of the human form, the solar system (with Earth displaced to indicate it is the planet from which the signal originated), and the transmitting radio telescope, with its diameter indicated.
If, (and that's a strong if) there are aliens out there, chances are that we'd receive a simple radio signal rather than intercept B'lorg's phone call to Vk'lar.
note to self (Score:2, Funny)
2. use their spare cycles to do my computing
3. ?
4. profit!
sorry, couldn't resist :)
the reason folding@home has more success (Score:2, Insightful)
I still prefer Seti@home, it looks cooler
Comment removed (Score:3, Insightful)
Re:The results were off by 20 percent! (Score:3, Interesting)
Also, keep in mind that this is a microsecond folding rate. Rates in the ms regime of folding are routinely measured with high accuracy, the microsecond regime is really hard as it usually takes longer than that to do what ever you're going to do to trigger folding in the first place. The number in the paper is 7.5us +/- 3.5us, so they got as close as could be expected.
folding@home and genome@home (Score:2, Informative)
Of course since glibc 2.3.1 killed the folding@home client....
For all of you wondering why... (Score:5, Interesting)
Distributed computing helps me sleep at night.
I've said it on K5 and I'll say it here. (Score:2, Informative)
Re:I've said it on K5 and I'll say it here. (Score:2, Interesting)
Re:I've said it on K5 and I'll say it here. (Score:3, Interesting)
Congratulations on getting slashdotted.
Re:I've said it on K5 and I'll say it here. (Score:5, Insightful)
render-not@home (Score:3, Interesting)
This could help all the hobbyist 3D movie makers a lot...
oh, yeah, and do some intelligent wordplay on not-at-home, render-nod, render-knot too
obscure, non-open-source, possibly commercial... (Score:4, Insightful)
How come everyone screams about software installed where nobody knows exactly what it does, the linux guys cry "HA-HA spy-ware cant happen to us" when they read about kazaa, and generally shun any binary-only software, but these things are cool and you have to "participate"?
Who sais these proggies really do what they claim?
Who tells you the guys providing the software don't just exploit you and will sell any results commercially as soon as they get some?
NSA employee 1: What could I do to get promoted?
NSA employee 2: (stumbles about distributed.net)
hey, look! great idea! we could to that too!
NSA employee 1: But How?
NSA employee 2: Hmm, don't those geeks freak out on Star Trek and Lego and stuff?
Re:obscure, non-open-source, possibly commercial.. (Score:2)
Their lusers also spam newsgroups boasting about it - with much the same spiel, talking about the "United Power of Linux" - whatever that might be, as the download page has versions for Windoze and MacOS as well...
And the lack of source is totally prohibitive here. No way on this planet am I running something like this on any box of mine.
Re:obscure, non-open-source, possibly commercial.. (Score:5, Insightful)
I never liked the fact that it was closed. their FAQ [berkeley.edu] is completely bogus. The fact is that somebody did send them packets with false results, and there's nothing they can do to prevent that. It was good old security through obscurity, and it didn't work. Besides, science is to be in the open.
However, what happened is that some client source code was left by a bunch of guys at Compaq, and as it happened, the optimized binaries they compiled (but did not distribute) ran 6 times as fast as the binaries provided by the project. We could complete a work unit in one hour if allowed to compile.
Even though the binaries must be approved by the project, there is no reason to keep people from compiling. On the contrary, you get optimized builds for each architecture. Not only that, you make use of the competence people have in writing more efficient code. Those who want to screw up their database can still do it with closed source, so there is absolutely not gaining anything.
It became too apparent that they didn't need my cycles, so I, and the rest of the group stopped. So, I know for sure that they lost one of their Top 50 Universities because of closed source.
I would happily run stuff on my box if it is done in the open, and the project management is responsive to enhancement requests. It doesn't need to be Free as in speech, or OSI approved, you just need to allow people to modify the source and contribute back to the project.
Sigh... IP? Anyone? (Score:5, Insightful)
Unlike other distributed computing projects, Folding@home is run by an academic institution (specifically the Pande Group, at Stanford University's Chemistry Department), which is a non-profit institution dedicated to science research and education. We will not sell the data or make any money off of it.
Ok, they won't make any money off it, but who might? Who owns any patents? What actually is done with the data? And the non-profit bit tells me nothing. The Vanguard Group [vanguard.com] is a non-profit too, but that doesn't mean they aren't interested in money. (Vanguard is owned by the investors, hence non-profit, but not really) Just because it is a non-profit institution doesn't tell me much. Universities are non-profit but they make a ton of money off of IP. They can do whatever they want but before I commit my processor cycles to helping I'd like to know specifically what I'm helping.
The FAQ goes on to say:
Moreover, we will make the data available for others to use. In particular, the results from Folding@home will be made available on several levels. Most importantly, analysis of the simulations will be submitted to scientific journals for publication, and these journal articles will be posted on the web page after publication. Next, after publication of these scientific articles which analyze the data, the raw data of the folding runs will be available for everyone, including other researchers, here on this web site.
So the data is going to be available. How? What "levels"? To whom? For how much? Just saying it will be published in journals tells me little. What else will be done with it? Who stands to benefit from the data? (aside from the obvious)
Basically I want to know and am not impressed with their answers. I'd like some candor when it comes to something this important. With SETI@home, who really cares? That won't affect my life. Folding@home might.
Can I join someone's team? (Score:3, Funny)
Would anyone mind if I joined their team?
SETI: let's hope aliens don't have lawyers (Score:5, Funny)
Hmm, maybe they have 8 arms and tentacles, and they'll just bite the lawyer's head off when they disagree with them.
This makes no sense, (Score:2)
Who is this guy, and why does he make such an obviously false statement? Should this make me willing to trust these people and contribute to this process?
If it takes one CPU-day to do 1ns of folding simulation, then protein folding is difficult to simulate because it occurs over a (relatively) long, not short time. This should be obvious, and therefore either the statement is either a deliberate lie or a misquote.
Re:This makes no sense, (Score:3, Informative)
New bumper sticker... (Score:3, Funny)
Comprehensive list of distributed projects (Score:3, Informative)
Re:And so what if SETI did get a hit? (Score:4, Insightful)
Re:And so what if SETI did get a hit? (Score:3, Insightful)
Re:And so what if SETI did get a hit? (Score:5, Interesting)
Earth: How do we cure cancer/every disease we know of?
ET: Use *this*, but now that you're living forever, you have to worry about massive overpopulation.
Earth: How do we get off this planet?
ET: Use *this*, but now you have to worry about war between your planets.
Earth: How do we achieve peace?
ET: Use *this*, but now you're bored outta your minds.
M@
Re:And so what if SETI did get a hit? (Score:5, Funny)
Earth: Wow, a spike that may have possibly been generated by an intelligent life form millions of years ago...
ET: ???
Re:And so what if SETI did get a hit? (Score:2, Insightful)
Re:And so what if SETI did get a hit? (Score:4, Insightful)
Sure they can. That's not in question. But the theory behind the distributed clients is to avoid wasting CPU cycles and to do something useful.
The point of the OP was that SETI@Home (and, frankly, RC5 crack searches) are osteniably no better than having the CPU cycles spinning anyway. Projects like Folding@Home, Genome@Home, and UD Cancer Research can provide a real, proveable benefit in both the short and long term. Mathematical projects like GIMPS and prime number searches do so as well, although my personal opinion is that they're not as valuable.
Use your CPU cycles however you like. Hell, don't run a distributed project at all if you don't want to. All that's being asked is to consider how to actually use the spare cycles effectively if you're going to join a distributed project.
Re:And so what if SETI did get a hit? (Score:5, Insightful)
(You're still right about SETI, though. What a freakin' waste.)
Re:And so what if SETI did get a hit? (Score:2, Insightful)
And what about all the tens or hundreds of millions that aren't running anything?
Re:And so what if SETI did get a hit? (Score:2, Interesting)
Each time I quit and removed it from my system.
Buggy installs, buggy software, buggy server statistics, unoptimized code, and a direct quote from Vijay that said "I don't care" about those issues because he already had enough people running the client for his purposes. Or at least that's my opinion of and the feeling that I was left with.
Re:.exe? (Score:2, Informative)
Re:SETI@Home hasn't had a hit... (Score:2)