FR Folding@Home Project Update - Folding of Proteins in Nanotube Confinement

Journal of the American Chemical Society ^ | 02/07/2006 | Stanford University

[texas booster]

From a paper in the Journal of the American Chemical Society concerning nanotubes:

The folding of proteins in confined spaces is a ubiquitous theme in biological and biomaterial applications, including folding in chaperones and pores, nanotube-based drug delivery, and cotranslational folding of nascent peptides in the ribosomal exit tunnel. The role of confinement on peptide conformational equilibrium has thus gained much interest in recent years, and a natural first hypothesis to investigate is the role of confinement alone in protein conformational preferences.

...

Still, there exists a growing body of evidence to suggest that molecular water plays a role in the conformational preferences and assembly of biomolecular systems. For example, it has been suggested that the addition of crowding agents or chemical denaturants destabilizes proteins by affecting the structure of water. Moreover, the character of water in confined environments is expected to differ significantly from that in bulk.

...

What role does water play in the stability of confined proteins? To address this question, we have simulated a well-characterized 23-residue helical peptide inside six fully solvated single-walled CNTs with diameters ranging from 15 to 35 Å. For each tube modeled, 1000 independent molecular dynamics trajectories were started from the fully helical and extended states (Figures 1 and 2a). Simulations were performed on the Folding@Home distributed computing network as described previously using the AMBER- 99 helix-coil force field and the TIP3P water model in the NVT ensemble at the approximate midpoint temperature of 305 K (89F). With individual trajectories on the 100-300 ns time scale and an aggregate time exceeding 2.5 ms, our extensive sampling allows us to extract equilibrium thermodynamic data.

(Please click on the above link for the full paper on Nanotube Confinement Denatures Protein Helices)

[texas booster]

We have had the Free Republic folding team for about 2 1/2 years now, and have grown dramatically in the last year. This is some of the folding work in which we may have participated.

Talk of nanotubes to deliver medicines may be the future, but there is a lot of hard work learning how to make it happen. How to pack the medicine inside the tube, how a protein may fold inside a tube and the effect of water molecules on such a small scale will all need to be examined before an effective delivery system utilizing nanotubes can be implemented.

To those not familiar with the Folding@Home distributed project please read on. We are here to help you get started!

[texas booster]

We have reached #57 on the list of teams worldwide, FReepers are in the Top 1,000 of all folders and we are stomping the bejebers out of the DUmmies and Daily Kos.

A year ago we were just beginning to expand. Now we have about 170 regular contributors with nearly 1,050 computers using F@H as their screensaver.

Folding@Home makes a real difference in basic research into such diseases as Alheimer's, Parkinson's and BSE.

All it takes is a little free time to download the core, which runs in the background. Safe, clean and effective!

[Incorrigible]

w00t!

[texas booster]

We have reached #57 on the list of teams worldwide, FReepers are in the Top 1,000 of all folders and we are stomping the bejebers out of the DUmmies and Daily Kos.

A year ago we were just beginning to expand. Now we have about 170 regular contributors with nearly 1,050 computers using F@H as their screensaver.

Folding@Home makes a real difference in basic research into such diseases as Alheimer's, Parkinson's and BSE.

All it takes is a little free time to download the core, which runs in the background. Safe, clean and effective!

[texas booster]

We have reached #57 on the list of teams worldwide, FReepers are in the Top 1,000 of all folders and we are stomping the bejebers out of the DUmmies and Daily Kos.

A year ago we were just beginning to expand. Now we have about 170 regular contributors with nearly 1,050 computers using F@H as their screensaver.

Folding@Home makes a real difference in basic research into such diseases as Alheimer's, Parkinson's and BSE.

All it takes is a little free time to download the core, which runs in the background. Safe, clean and effective!

[texas booster]

As always, please ping me in you want on or off the list!

[Clara Lou]

Woo hoo! My computers are due to complete 3 units today! [Of course, that means it may be a while before any more are completed.]

[Drango]

Proud to b foldin' 4 the "Gipper"

[texas booster]

If you have ever folded a p2061, p2066, p2067, p2071, p2074, p2076, p2094, p2097 or a p2098 assignment then you participated in this work.

One day we will have nanotube delivery of medicines and genetics. This is the hard work plowing new ground that needs to be done.

[texas booster]

Folding@Home FAQ for new users:

What is Folding@Home? A Stanford University project to find out how proteins fold.

Why it's important: Proteins folding wrong causes all kinds of diseases, like Alzheimer's, Parkinson's, and forms of cancer. Folding@Home uses novel computational methods and large scale distributed computing, to simulate timescales thousands to millions of times longer than previously achieved. Through Folding@home, scientists now have the horsepower to study the mechanics of protein folding. With its ability to share the workload among hundred of thousands of computers economically, Folding@home can help scientists understand how proteins snap, or don't, into their predestined shapes - and may help to explain the origins of diseases such as Alzheimer's and apparently unrelated diseases. We're fueling research that could end all that.

How does it work?: You download a safe, tested program (see link below) that is certified by Stanford University. It gets work from Stanford, runs calculations using your spare computer power, and sends the results back to the University.

Is it safe? Yes! Folding@Home rarely effects computer performance in any way and won't compromise your privacy in any way. It only uses the computing power you aren't using so it doesn't slow down other programs.

How do I get started folding for Team FreeRepublic?:
1.) Download the folding program from Stanford University's folding download page (Folding@home Client Download). Type in your desired username.
2.) Type in 36120 for the team number. THIS IS VERY IMPORTANT - if you get the number wrong, you won't be folding for team FreeRepublic!
3.) The third question asks, "Launch automatically at machine startup, installing this as a service?" - We recommend you answer YES. Otherwise you will have to manually start the program after every reboot.

How can my computer help? Even if they were given exclusive access to all of the world's supercomputers, Stanford still wouldn't have as much processing power as they get from the supercluster of people's desktop systems Folding@home relies on. Modern supercomputers are essentially a cluster of hundreds of processors linked by fast networking. But Stanford needed the power of hundreds of thousands of processors, not just hundreds.

There's no reason to not get involved! It's free, easy, and you can know you're helping every minute without lifting a finger.

*******************************************

List of Relevant Folding Links
Why Fold - Watch This !!

Another Folding Clip

Folding@home Client Download

FreeRepublic.com Folder Stats

Extreme Overclockers Stats for FreeRepublic

Another Stats Page

*******************************************
Competition (Not!!) Dummies ..Daily Kos

Dummie Folding Threads #7 #8 #9#10#11 #12

**************************************************
Other Useful Stuff - Links

How much are those work units worth? And what are they?
All Projects Listed
Point Summary for Workunits

Stat Image Generator

Fahmon Third Party Monitoring Software

**************************************
Past FreeRepublic Folding threads

#1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #17 #18 #19 #20#21#22 #23 #24 #25 #26 #27 #28

[E.G.C.]

BTTT

[texas booster]

Another completed project, that provided results that we now build upon. How the molecules G5, C6, A7 and A8 fold into itself yet remain active was the object of many years of folding.

The GNRA hairpin is a small secondary structural unit of RNA that contains what some have called an "unusually stable" tetraloop region bridging the two strands that form this helical hairpin. We have previously characterized short-timescale loop dynamics and thermal unfolding of the hairpin and have since undertaking numerous related studies in silico including: unfolding at a biologically relevant temperature, direct folding, evaluation of the equilibrium thermodynamics of this RNA hairpin via replica exchange stochastic dynamics, and calculations of the folding probabilities for a small ensemble of structures (Pfold), which allow us to examine the transition state for the folding event.

This series of projects is now complete, and is the launching point for the new 340-385 series which uses much more sophisticated algorithms to maximize the accuracy obtained and to examine the role of water molecules and ions in the folding and unfolding processes.

[Paladin2]

At #57, the climb is steep.

[LonePalm]

According to my spreadsheet we will pass the following teams:

AoA Forum: 1/30/07
MozillaZine: 4/3/07
Linux: 4/9/07
Team 2ch: 6/27/07

Unfortunately we get passed by Team xtremesystems.org on 9/27/07.

The good news is that these dates are heavily weighted against us because it covers Christmas-New Year's week when many of Klutz_dohanger's servers are down.

Garde la Foi, mes amis! Nous nous sommes les sauveurs de la République! Maintenant et Toujours!
(Keep the Faith, my friends! We are the saviors of the Republic! Now and Forever!)
LonePalm, le Républicain du verre cassé (The Broken Glass Republican)

[processing please hold]

Thanks for the new thread.

My head hurts, I gotta get me some aspirins. :(

[texas booster]

xstremesystems.org had a contest to get to the 10 million mark. They are a high powered team with a lot of systems. Now their top folders are moving back to the World Computing Grid. We will be OK in the long run.

There are only 56 teams for us to pass, and most have been around a very long time. Our point totals will continue to increase and keep us competitive.

If anyone has upgraded a system this Christmas, or gotten an ATI X-series video card let us know so we can track your increase in points and maximize your production.

Egon's tool will make following your F@H points much easier. FReepmail Egon for the links to start.

[brityank]

Thanks, texas booster, for letting me help. my little WinME is running ~15k a year to the pot.

I've also found that if I shut down F@H before I reboot (updates, etc.) then I don't lose the points generated up to then. Power fails -- bye-bye!

Happy New Year, All.

[Route66]

Please Join the FR Team Bump

[texas booster]

Shutting down the 4.0 GUI version will save a check point in the log so that you can take off from where you last folded.

In case of a system shutdown I would think that you would only lose the points since the last check point, not the results from the entire fold. Let me look into that.

See if you can set the GUI to save a check point every 5 minutes or so.

[Coleus]

you don't even know it is there, just sign up and put in the FR code and your off, helping science and humanity.

[LonePalm]

My ATI card will probably arrive on Wednesday. Our company follows DoD holidays and closings.

Garde la Foi, mes amis! Nous nous sommes les sauveurs de la République! Maintenant et Toujours!
(Keep the Faith, my friends! We are the saviors of the Republic! Now and Forever!)
LonePalm, le Républicain du verre cassé (The Broken Glass Republican)

[texas booster]

Thank you for joining us and publicising the team. Have you been folding very long?

For any of the new folders please pop a note here if you have any problems.

TB

[brityank]

Thanks; my checkpoint is set for 15Min, but it only seems to do them at each step anyway:

[12:36:37] + Working...
[13:34:40] Writing local files
[13:34:40] Completed 12200000 out of 20000000 steps (61)
[15:45:36] Writing local files
[15:45:36] Completed 12400000 out of 20000000 steps (62)
[18:01:26] Writing local files
[18:01:26] Completed 12600000 out of 20000000 steps (63)
[18:36:34] + Working...


Not a problem; I learned on my company's laptop not to allow the IT mavens to upload/reboot my machine after they clobbered five hours I had devoted into a spreadsheet.

Oh, yes. They did hear me now! ;^)

[ken in texas]

Thanks for the new thread.

My compute resources have subsided a bit since first starting, but I now have another machine that should be able to fold for 3-4 months.

[paulat]

Bump!

C'mon, you guys...join the team!!

I can't believe I don't see more familiar names on our list!!

[JosephW]

Things are moving along swimmingly and I should be among the top 7500 folders in another day. However, with the new year, Egon's tool is giving me some really funny results. Looks like most of my WU won't finish until sometime 2 or 3 years out (I know it's an error, but it is funny).

At this rate I'm not sure how much of a help I'm going to be until about 2010 :)

[processing please hold]

I can't believe I don't see more familiar names on our list!!

I've thought the same thing. Oh well...

[dfwddr]

At this rate I'm not sure how much of a help I'm going to be until about 2010 :)

The best solution for you would be to shut down, oh for somewhere around 4-6 weeks. Your climbing into the top 10 way too fast, and I think this would be the best thing for you. Trust me ;)

[JosephW]

Relax, it's at least 5 weeks until I catch up to you, probab ly more like 7.

Since I don't have any upgrades on the horizon, you can expect me to average about 800ppd

[texas booster]

I have noticed the same accounting on the logs, where it only makes an entry after a step. When I looked closer I found additional entries on a system where F@H is shut down regularly.

---

[18:43:19] Assembly optimizations on if available.
[18:43:19] Entering M.D.
[18:43:39] (Starting from checkpoint)
[18:43:39] Protein: p2126_villin_sigd3
[18:43:39]
[18:43:39] Writing local files
[18:43:39] Using table 3
[18:43:40] Completed 3188807 out of 50000000 steps (6)
[18:43:40] Extra SSE boost OK.
[19:48:45] Writing local files
[19:48:45] Completed 3500000 out of 50000000 steps (7)
[21:33:01] Writing local files
[21:33:01] Completed 4000000 out of 50000000 steps (8)

Folding@Home Client Shutdown.

---

At least on this system, Win XP SP2, F@H GUI 5.03 does record shut downs between steps.

You should have made the IT mavens (coven?) recover the file. Nowadays spreadsheets keep writing open files to a temp file, that if all goes well, can usually be recovered by a competent IT dude / dudette.

[texas booster]

Looking at your points chart, it seems that you have lost a system, or had one stop reporting into Stanford.

http://folding.extremeoverclocking.com/user_summary.php?s=&u=197716

Once the ATI card is in, install the recommended drivers as listed here:

http://www.stanford.edu/group/pandegroup/folding/FAQ-ATI.html

Once installed, test F@H by looking at the logs. A sure sign of driver mismatch is completing a WU every minute - they are actually abnormal ends (abends) and do not contribute to actual research. One fella on another team recorded 202 points on 1,448 work units.

You should be able to complete a WU every day, boosting your ppd to 600 - 700 on just that machine. Add in a couple of other systems and you could easily be a 1,000 ppd producer.

Plus you can play any game made at full resolution.

[LonePalm]

Well, those charts are certainly better than my little, manually updated, POS spreadsheet.

Garde la Foi, mes amis! Nous nous sommes les sauveurs de la République! Maintenant et Toujours!
(Keep the Faith, my friends! We are the saviors of the Republic! Now and Forever!)
LonePalm, le Républicain du verre cassé (The Broken Glass Republican)

[Egon]

However, with the new year, Egon's tool is giving me some really funny results. Looks like most of my WU won't finish until sometime 2 or 3 years out (I know it's an error, but it is funny).

Yeah, that's a new feature I've added: REALLY long-term predictions.

F@H doesn't report years with their dates in the status file. Apparently, I failed to take that into consideration. Working on a fix now.

[Egon]

My shameless self-promotion for this thread:

If you're interested in tracking your folding machine(s) over the web, please Freepmail me.

Available features include:

• Timely snapshot of each machine's progress on it's current project
• Point values for project
• Estimated completion dates/times
• Warning indicators for machines that appear to have stopped communicating or folding
• Warnings for machines that won't make their deadline
• Team ranking status box (courtesy of EOC)
• Machine stats like PPD (points per day)
• Project comparison stats: PPD, average time taken to completion, etc.
• It's free-- and always will be
• ...other features as time permits, and people request

[Egon]

OK. Problem fixed, I think.

I'm going back through and looking for weird history entries turned in over the last day, and I'll attempt to correct them as well.

If, over the next couple of days, you notice any weird history entries, let me know the machine, date, and project, and I'll correct them.

Thanks!

[brityank]

Yes; that's a normal shutdown:

Folding@home Client Shutdown.


--- Opening Log file [December 6 04:57:54]


# Windows Graphical Edition ####################

Here's what a failed one gives:

[12:43:27] Writing local files
[12:43:27] Completed 495000 out of 750000 steps (66)


--- Opening Log file [December 6 00:36:00]


# Windows Graphical Edition ################

There's no shutdown, and the files are corrupted so it restarts from 0. I also find that the SSE Boost is missing, but a clean restart gets it back.

[texas booster]

I looked at this issue many months ago and needed a refresh. The info on recovering defective work units can be found here:

http://fahwiki.net/index.php/Howto_use_qd_to_debug_the_FAH_client

It may be wiki but it seems clean.

Sometime work units will calculate into improbable folding situations and will be killed off. Sometimes, what you seem to experience happens. I'll bet that your client databases are scrambled and need repair.

The qd utility refrenced above in wiki will sometimes fix queue.dat and allow you to recover your points.

Disclaimer: I have never taken the time to learn this utility and will not be able to for a week or so. However, if my 9:00 pm contribution is any indication (2 work units for 11 points) then I may need to learn it well.

[HangThemHigh]

My power company sees fit to provide power here at the house most of the time. A year or so ago, if I lost power or even a brief flicker, I had about a 30% chance of losing the WU in progress. The checkpoint file seemed to be getting corrupted in a power loss. It got bad enough that I wrote a script to back up the whole directory about once an hour to preclude massive losses.

Lately Stanford seems to have fixed the console version so that on lost power the previously saved checkpoint isn't corrupted. I'm assuming they have "fixed" this as I haven't had to restore the F@H directory from my backups in 6 mo. or so.

Perhaps the GUI version hasn't gotten the fix, or Windows ME (I run XP) prevents the fix from working somehow. Even before the fix, if I shut down the computer normally (vs. a power failure or hitting the reset button) the checkpoint file would survive.

Not sure this helps, just an observation.

[Egon]

Perhaps the GUI version hasn't gotten the fix, or Windows ME (I run XP) prevents the fix from working somehow.

I think it might be Windows ME, actually. I've had (until recently) a WinME machine folding with the console version. Every once in awhile, it would lock up, and upon restarting, the WU would invariably restart from scratch.

I haven't had this problem on any other OS.

[processing please hold]

[07:57:41] Project: 3301 (Run 3898, Clone 0, Gen 1)
[07:57:41]
[07:57:41] Assembly optimizations on if available.
[07:57:41] Entering M.D.
[07:57:48] Protein: p3301_ribcomp
[07:57:48]
[07:57:48] Writing local files
[07:57:49] Extra SSE2 boost OK.
[07:57:50] Writing local files
[07:57:50] Completed 0 out of 25000 steps (0)
[08:04:26] Writing local files

[12:35:30] Writing local files
[12:35:31] Completed 21500 out of 25000 steps (86)
[12:37:56] Quit 101 - Fatal error: Box exploding.
[12:37:56]
[12:37:56] Simulation instability has been encountered. The run has entered a
[12:37:56] state from which no further progress can be made.
[12:37:56] This may be the correct result of the simulation, however if you
[12:37:56] often see other project units terminating early like this
[12:37:56] too, you may wish to check the stability of your computer (issues
[12:37:56] such as high temperature, overclocking, etc.).
[12:37:56] Going to send back what have done.
[12:37:56] logfile size: 8556
[12:37:56] - Writing 9109 bytes of core data to disk...
[12:37:56] ... Done.
[12:37:56]
[12:37:56] Folding@home Core Shutdown: EARLY_UNIT_END
[12:38:00] CoreStatus = 72 (114)
[12:38:00] Sending work to server

[12:38:00] + Attempting to send results
[12:38:03] + Results successfully sent
[12:38:03] Thank you for your contribution to Folding@Home.
[12:38:07] - Preparing to get new work unit...
[12:38:07] + Attempting to get work packet
[12:38:07] - Connecting to assignment server
[12:38:08] - Successful: assigned to (171.64.65.83).
[12:38:08] + News From Folding@Home: Welcome to Folding@Home
[12:38:08] Loaded queue successfully.
[12:45:39] + Closed connections
[12:45:44] [12:45:44] + Processing work unit
[12:45:44] Core required: FahCore_79.exe
[12:45:44] Core found.
[12:45:44] Working on Unit 07 [January 3 12:45:44]
[12:45:44] + Working ...
[12:45:44]
[12:45:44] *------------------------------*
[12:45:44] Folding@Home Double Gromacs Core
[12:45:44] Version 1.91 (April 11, 2006)
[12:45:44]
[12:45:44] Preparing to commence simulation
[12:45:44] - Looking at optimizations...
[12:45:44] - Created dyn
[12:45:44] - Files status OK
[12:45:45] - Expanded 1381063 -> 5334069 (decompressed 386.2 percent)
[12:45:45] - Starting from initial work packet
[12:45:45]
[12:45:45] Project: 3301 (Run 3645, Clone 0, Gen 1)
[12:45:45]
[12:45:46] Assembly optimizations on if available.
[12:45:46] Entering M.D.
[12:45:52] Protein: p3301_ribcomp
[12:45:52]
[12:45:52] Writing local files
[12:45:53] Extra SSE2 boost OK.
[12:45:55] Writing local files
[12:45:55] Completed 0 out of 25000 steps (0)
[14:00:41] Writing local files
[14:00:41] Completed 6000 out of 25000 steps (24)
[14:08:30] Quit 101 - Fatal error: Box exploding.
[14:08:30]
[14:08:30] Simulation instability has been encountered. The run has entered a
[14:08:30] state from which no further progress can be made.
[14:08:30] This may be the correct result of the simulation, however if you
[14:08:30] often see other project units terminating early like this
[14:08:30] too, you may wish to check the stability of your computer (issues
[14:08:30] such as high temperature, overclocking, etc.).
[14:08:30] Going to send back what have done.
[14:08:30] logfile size: 8454
[14:08:30] - Writing 9007 bytes of core data to disk...
[14:08:30] ... Done.
[14:08:31]
[14:08:31] Folding@home Core Shutdown: EARLY_UNIT_END
[14:08:33] CoreStatus = 72 (114)
[14:08:33] Sending work to server

Ok, there was one more that did the same thing, I didn't post it because too much space. I deleted the steps between the first and last numbered steps to save space. By looking at it can you tell what happened?

[texas booster]

Yes I can tell what happened.

Your box exploded.

Actually it looks like the Double Gromacs core that this work units uses encountered an out-of-range error and gave you an Early_Unit_End. I have had several of these, all on p33xx work units.

Looking at your points charts I see that you were credited for two work units today, one for 19 points and one for 40 points. That is a sure sign of a EUE caused by work unit errors.

The server seems to connect OK, and there do not seem to be any other issues with overclocking. It is possible that the SSE boost is the source of the problem but I haven't been able to confirm it.

Stanford will tighten up the boundaries of these work units within a week, once they discover the source of the EUEs.

[texas booster]

http://folding.extremeoverclocking.com/user_summary.php?s=&u=168262

Link to your daily points, with a pretty graph.

[processing please hold]

My box exploded...thank God I didn't get any glass in my eyes from the explosion, huh? :-)

It's given me another p3301, we'll see if I'm lucky enough not to be injured by more flying glass. Must be a problem with their p33xx's

My p2124 is on step 96 and should be through soon.

My SSE boost shows to be ok.

[processing please hold]

I like that. It is pretty. Bookmarked. Thanks.

[NY.SS-Bar9]

Almost done building my new rig - AMD3800+X2 / ATI X1900

Should smoke those little proteins.

[texas booster]

Will the MB have 2 CPUs? The new SMP betas are pumping out about 1,000 ppd for every system. Add in the GPU clients (not sure if they will work together), for 600 ppd, you could have a real barn burner.

The GPU can run as fast as possible, will use a moderate amount of CPU just for housekeeping, and the CPU probably can run a second core at about 25% - 40%.

Just be sure not to load Office, any games or other useful stuff. You are spending all this money for F@H. ;)

[texas booster]

Congrats on your first completed work unit for F@H.

http://folding.extremeoverclocking.com/user_summary.php?s=&u=205119

Ping us if you have any questions!

TB

[antiRepublicrat]

Congrats on your first completed work unit for F@H.

Thanks! I was going to write in and brag, but you beat me to it. I can't put much processor towards it, but every little bit helps. There'll be some more coming if I can get a certain notebook power connector fixed.

[antiRepublicrat]

My power company sees fit to provide power here at the house most of the time. A year or so ago, if I lost power or even a brief flicker, I had about a 30% chance of losing the WU in progress.

My UPS has saved my butt six times so far according to the software.

[Mount Athos]

I got a new dual core CPU installed and running with a folding process assigned to each one.