Rosetta is (we hope) ready for prime time

Author	Message
David Baker Send message Joined: 5 Nov 05 Posts: 7 Credit: 1,000 RAC: 0	Message 186019 - Posted: 5 Nov 2005, 6:23:43 UTC Hello All! I'm the lead scientist at rosetta@home, and would like to encourage some of you to join on part time. We need all the computing power we can get! While we have made great progress in the last month with our BOINC project (which would not have been possible without all the great work that David Anderson and others here have done), it is clear that to consistently predict protein structure accurately requires still more compute power. I'd be happy to answer any questions you may have about the project here. there are some articles about what we are doing at http://boinc.bakerlab.org/rosetta/rah_articles.php hope to see some of you at rosetta@home! David ID: 186019 ·

UBT - Halifax--lad Volunteer tester Send message Joined: 13 Dec 00 Posts: 433 Credit: 13,900 RAC: 0	Message 186057 - Posted: 5 Nov 2005, 10:22:13 UTC I would have to agree that it is ready for prime time it is like any other BOINC project it has its ups and downs just as SETI did, to work out these problems they need users to crunch work and report any problems if any. So long as you have a decent computer with 512Mb of ram or more you should be fine, some users with less have experienced problems in the past I did but the problems have gone now. Join us in Chat (see the forum) Click the Sig Join UBT ID: 186057 ·

Paul D. Buck Volunteer tester Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0	Message 186063 - Posted: 5 Nov 2005, 11:26:54 UTC There has not been, and likely will not be, a project that does not have problems at one time or another. Rosetta@Home has had a couple issues recently, but so has Predictor@Home, and CPDN, and SETI@Home ... More good news is that the project is trying hard to stay on top of the problems ... just like all the other projects. Of course, I am prejudiced ... :) ID: 186063 ·

Al Volunteer tester Send message Joined: 4 Oct 99 Posts: 5832 Credit: 401,935 RAC: 0	Message 186066 - Posted: 5 Nov 2005, 12:57:01 UTC - in response to Message 186019. What is your goal?(wich deseases do you whant to cure) I hope this does not sound dumb:) Scorpions - Wind Of Change ID: 186066 ·

Andrew Volunteer tester Send message Joined: 20 May 99 Posts: 40 Credit: 29,416 RAC: 0	Message 186101 - Posted: 5 Nov 2005, 17:32:40 UTC Last modified: 5 Nov 2005, 17:33:02 UTC Here are some additional links for those who haven't looked around Rosetta's website: Rosetta's Homepage What is Rosetta What they are doing How they are doing it Interesting Articles Recent paper - Progress in Modeling of Protein Structures and Interactions (pdf) ID: 186101 ·

David Baker Send message Joined: 5 Nov 05 Posts: 7 Credit: 1,000 RAC: 0	Message 186109 - Posted: 5 Nov 2005, 17:58:01 UTC - in response to Message 186066. What is your goal?(wich deseases do you whant to cure) I hope this does not sound dumb:) No, it doesn't sound dumb! Currently our structure prediction work is mainly focusing on developing and testing improved algorithms (I give feedback on our message boards about what we are learning every week or so), but when we can predict structure more accurately the method should contribute to finding cures for many diseases. We are also working on designing new HIV vaccines, and we will be running these calculations at rosetta@home soon as well. ID: 186109 ·

Siran d'Vel'nahr Volunteer tester Send message Joined: 23 May 99 Posts: 7381 Credit: 44,181,323 RAC: 238	Message 186113 - Posted: 5 Nov 2005, 18:07:28 UTC - in response to Message 186019. Hello All! I'm the lead scientist at rosetta@home, and would like to encourage some of you to join on part time. We need all the computing power we can get! While we have made great progress in the last month with our BOINC project (which would not have been possible without all the great work that David Anderson and others here have done), it is clear that to consistently predict protein structure accurately requires still more compute power. I'd be happy to answer any questions you may have about the project here. there are some articles about what we are doing at http://boinc.bakerlab.org/rosetta/rah_articles.php hope to see some of you at rosetta@home! David David, I placed your message above on the front page of my website. Hopefully, I can help you spread the word. I have a link to Rosetta@Home and 15 other BOINC projects, plus the official BOINC site on mine. My site link is in my sig below. Good luck with your project.... >:-) (-:< CAPT Siran d'Vel'nahr - L L & P _\\// Winders 11 OS? "What a piece of junk!" - L. Skywalker "Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath ID: 186113 ·

Al Volunteer tester Send message Joined: 4 Oct 99 Posts: 5832 Credit: 401,935 RAC: 0	Message 186145 - Posted: 5 Nov 2005, 20:41:08 UTC - in response to Message 186109. Hi David just attached to R@H:) Im gona help out:) Scorpions - Wind Of Change ID: 186145 ·

David Baker Send message Joined: 5 Nov 05 Posts: 7 Credit: 1,000 RAC: 0	Message 186241 - Posted: 6 Nov 2005, 6:27:34 UTC - in response to Message 186113. Hello All! I'm the lead scientist at rosetta@home, and would like to encourage some of you to join on part time. We need all the computing power we can get! While we have made great progress in the last month with our BOINC project (which would not have been possible without all the great work that David Anderson and others here have done), it is clear that to consistently predict protein structure accurately requires still more compute power. I'd be happy to answer any questions you may have about the project here. there are some articles about what we are doing at http://boinc.bakerlab.org/rosetta/rah_articles.php hope to see some of you at rosetta@home! David David, I placed your message above on the front page of my website. Hopefully, I can help you spread the word. I have a link to Rosetta@Home and 15 other BOINC projects, plus the official BOINC site on mine. My site link is in my sig below. Good luck with your project.... >:-) (-:< Thank you, Siran! ID: 186241 ·

Siran d'Vel'nahr Volunteer tester Send message Joined: 23 May 99 Posts: 7381 Credit: 44,181,323 RAC: 238	Message 186493 - Posted: 7 Nov 2005, 0:10:39 UTC - in response to Message 186241. Hello All! I'm the lead scientist at rosetta@home, and would like to encourage some of you to join on part time. We need all the computing power we can get! While we have made great progress in the last month with our BOINC project (which would not have been possible without all the great work that David Anderson and others here have done), it is clear that to consistently predict protein structure accurately requires still more compute power. I'd be happy to answer any questions you may have about the project here. there are some articles about what we are doing at http://boinc.bakerlab.org/rosetta/rah_articles.php hope to see some of you at rosetta@home! David .... Thank you, Siran! No problem David, my pleasure, you're quite welcome. My site is my small contribution to the world of BOINC. >:-) (-:< CAPT Siran d'Vel'nahr - L L & P _\\// Winders 11 OS? "What a piece of junk!" - L. Skywalker "Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath ID: 186493 ·

Misfit Volunteer tester Send message Joined: 21 Jun 01 Posts: 21804 Credit: 2,815,091 RAC: 0	Message 186500 - Posted: 7 Nov 2005, 0:36:23 UTC - in response to Message 186066. What is your goal?(wich deseases do you whant to cure) Other than Rosetta@home you also have... Protein folding: Folding@home Predictor@home World Community Grid Finding drugs: Drug Design (infectious diseases) FightAIDS@home (HIV/AIDS) National Foundation (cancer) me@rescam.org ID: 186500 ·

Crunch3r Volunteer tester Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0	Message 186503 - Posted: 7 Nov 2005, 0:44:59 UTC - in response to Message 186109. What is your goal?(wich deseases do you whant to cure) I hope this does not sound dumb:) No, it doesn't sound dumb! Currently our structure prediction work is mainly focusing on developing and testing improved algorithms (I give feedback on our message boards about what we are learning every week or so), but when we can predict structure more accurately the method should contribute to finding cures for many diseases. We are also working on designing new HIV vaccines, and we will be running these calculations at rosetta@home soon as well. Will the source code be released to public ? Join BOINC United now! ID: 186503 ·

EclipseHA Send message Joined: 28 Jul 99 Posts: 1018 Credit: 530,719 RAC: 0	Message 186512 - Posted: 7 Nov 2005, 1:10:15 UTC - in response to Message 186503. Will the source code be released to public ? I hope not! Seti is the only project that does, and I think that there's a very valid "science" reason not to do so! ID: 186512 ·

ML1 Volunteer moderator Volunteer tester Send message Joined: 25 Nov 01 Posts: 21725 Credit: 7,508,002 RAC: 20	Message 186520 - Posted: 7 Nov 2005, 1:25:04 UTC - in response to Message 186512. Last modified: 7 Nov 2005, 1:25:27 UTC Will the source code be released to public ? I hope not! Seti is the only project that does, and I think that there's a very valid "science" reason not to do so! AZ: Please explain why? Regards, Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) ID: 186520 ·

David Baker Send message Joined: 5 Nov 05 Posts: 7 Credit: 1,000 RAC: 0	Message 186552 - Posted: 7 Nov 2005, 4:00:59 UTC - in response to Message 186520. Last modified: 7 Nov 2005, 4:02:48 UTC Will the source code be released to public ? I hope not! Seti is the only project that does, and I think that there's a very valid "science" reason not to do so! AZ: Please explain why? Regards, Martin Yes. we will be releasing the source code this week. This will make it possible for people to optimize compilation for their own machines, and for platforms we aren't able to test on because we don't have them in house. Experts who want to get more involved can also work on tuning the code and experimenting with search algorithms (which is what I spend most of my research time doing). The rosetta code is already used by researchers around the world, and our philosophy is that the more people working with and using the code the faster it will improve. The code is free to non corporate users, but companies have to pay about 25K/yr for a license--this money supports an annual developers meeting which all people working with the code are invited to attend (all expenses paid). decisions about what direction to take with the code for the next year are made by consensus at this meeting. we had over 60 people last year and it was great fun. if any of you get involved seriously you can come next summer--it is in the washington cascades. on a different topic, I think I made something of a breakthrough staring at the rosetta@home results today. more on this on the rosetta@home message boards in the next few days. ID: 186552 ·

Paul D. Buck Volunteer tester Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0	Message 186602 - Posted: 7 Nov 2005, 9:56:07 UTC Martin, With the source code open and available for modification, well, the unscrupulous amongst us may do just that ... make modifications that compromise the science. Though we have some mechanisms in place to limit the damage, I still question if they are strong enough to provide adequate safeguards. The problem is that the rule of reciprocals can come into play ... Add a teaspoon of wine to sewage and you get sewage ... and if you add a teaspoon of sewage to a barrel of wine and you get sewage ... This to me is especially worrysome with a project like Rosetta@Home that is not currently using redundency. Though, if I understand the current state of the project, we are not in a mode where bad results really could hurt the project. Use of "voting" is fine, except in those cases where the two or three bad results locks out the one good result (happened on a space launch where 2 computers with the wrong answer out-voted the computer with the right answer). ID: 186602 ·

Al Volunteer tester Send message Joined: 4 Oct 99 Posts: 5832 Credit: 401,935 RAC: 0	Message 186682 - Posted: 7 Nov 2005, 16:39:37 UTC - in response to Message 186500. What is your goal?(wich deseases do you whant to cure) Other than Rosetta@home you also have... Protein folding: Folding@home Predictor@home World Community Grid Finding drugs: Drug Design (infectious diseases) FightAIDS@home (HIV/AIDS) National Foundation (cancer) Thanks Misfit il chek it out:) Scorpions - Wind Of Change ID: 186682 ·

Fuzzy Hollynoodles Volunteer tester Send message Joined: 3 Apr 99 Posts: 9659 Credit: 251,998 RAC: 0	Message 186813 - Posted: 8 Nov 2005, 0:41:34 UTC - in response to Message 186602. Last modified: 8 Nov 2005, 0:41:55 UTC Martin, With the source code open and available for modification, well, the unscrupulous amongst us may do just that ... make modifications that compromise the science. Though we have some mechanisms in place to limit the damage, I still question if they are strong enough to provide adequate safeguards. ...... Paul, I share your concern as Rosetta only requires one WU with a valid result. So sending WU's to more than one computer will be a good idea, so a quorum can be made and the results can be compared. "I'm trying to maintain a shred of dignity in this world." - Me ID: 186813 ·

David Baker Send message Joined: 5 Nov 05 Posts: 7 Credit: 1,000 RAC: 0	Message 186888 - Posted: 8 Nov 2005, 5:49:10 UTC Redundancy seems wasteful since we are always limited in what we can solve by available computing power, but we will resort to this if necessary. So far we have not had any problems, and we can easily recompute energies in house to double check the lowest energy structures found by the community. ID: 186888 ·

Tern Volunteer tester Send message Joined: 4 Dec 03 Posts: 1122 Credit: 13,376,822 RAC: 44	Message 186902 - Posted: 8 Nov 2005, 7:27:48 UTC - in response to Message 186888. Redundancy seems wasteful Can you use the flop-counting available in V5.x? I understand that the redundancy isn't needed for your application's science, as anyone who cheats is unlikely to come up with the lowest energy, or if they do, it's easy to check if they really did it. But the credit issue is possibly a problem. At the moment, granted credit = claimed credit. With redundancy, you drop the highest and lowest, so cheating on benchmarks doesn't gain all that much. Without, cheating-for-credit could become a problem. Counting flops rather than benchmark*time would maybe cost a few percent of the available resource, as opposed to 66% if you issued each WU 3 times... ID: 186902 ·

©2025 University of California

SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.