Rosetta@home

CASP9

  UW Seal
 
[ Home ] [ Join ] [ About ] [ Participants ] [ Community ] [ Statistics ]
  [ login/out ]


Advanced search
Message boards : Rosetta@home Science : CASP9

Sort
AuthorMessage
Mike Tyka

Joined: Oct 20 05
Posts: 96
ID: 5612
Credit: 2,190
RAC: 0
Message 65811 - Posted 25 Apr 2010 1:04:17 UTC
Last modified: 25 Apr 2010 1:14:10 UTC

Hey everyone !

As some of you may know CASP9, a community wide experiment in structure prediction, is starting in less then 10 days. As usual we are participating with both a human prediction group and a brand new automatic server, called RosettaServer. We will be utilizing BOINC heavily during the next 4 months, as you can see we already have nearly 4 million workunits lined up.
We have been very busy getting everything ready in time!

CASP 9
CASP is an international experiment to assess the state-of-the-art of the protein structure prediction field. Sequences, whose structures have been solved but which have not yet been published are sent out to participating teams and we have a limited amount of time to send back predictions. The whole thing is conducted in a double-blind fashion ensuring fair assessment and truly blind prediction.

We will be testing a number of new strategies during this period, these will be carried out by the human prediction group and crunched on BOINC.

In the automatic category we have modernized out old prediction server (Robetta, created more then 6 years ago) and replaced it's heart with Rosetta's most modern implementation. Among many scientific improvements, we have also, for the first time, connected the public structure prediction server to BOINC for both ab initio prediction and homology modeling.

For the community, by the community:
After CASP this server will go public and provide Rosetta's Abinitio and Homology modelling capabilities to the scientific community (and to the public). The computation for this will be conducted on BOINC meaning that you guys will be crunching protein structure prediction jobs for real scientific studies conducted by researchers all over the world!


Thanks again everyone for crunching, we wouldn't be able to do this stuff without you !

Excitedly yours,

Mike
____________
http://beautifulproteins.blogspot.com/
http://www.miketyka.com/

Chilean Profile
Avatar

Joined: Oct 16 05
Posts: 651
ID: 5008
Credit: 10,238,180
RAC: 4,709
Message 65812 - Posted 25 Apr 2010 1:41:22 UTC

Awesome! Winter just started here... Meaning my PCs will be both crunching and making the winter a bit less cold. :)
____________

Michael G.R.

Joined: Nov 11 05
Posts: 263
ID: 11128
Credit: 8,342,700
RAC: 5,198
Message 65822 - Posted 25 Apr 2010 18:46:09 UTC
Last modified: 25 Apr 2010 18:46:35 UTC

Great, I'll try to throw more computing power your way for the duration! It will be interesting to see how Rosetta has improved since the last CASP.

btw, this announcement should probably linked from the frontpage news.
____________

Matthew Lei
Avatar

Joined: Jun 5 06
Posts: 4
ID: 87065
Credit: 258,058
RAC: 0
Message 65847 - Posted 27 Apr 2010 3:09:04 UTC

Guess I'll relocate more CPU time for this event. :)
____________

JChojnacki Profile
Avatar

Joined: Sep 17 05
Posts: 71
ID: 105
Credit: 6,746,778
RAC: 1,867
Message 65848 - Posted 27 Apr 2010 3:53:42 UTC

Very excited about CASP9 this year.

And Mike, a double "WOW" goes out to you all for the "For the community, by the community" section of your post. That is very cool news!

Happy CASP9-ing.

~Joel
____________



Matthew Lei
Avatar

Joined: Jun 5 06
Posts: 4
ID: 87065
Credit: 258,058
RAC: 0
Message 65886 - Posted 29 Apr 2010 7:51:12 UTC

Just wondering, it is possible tell whether the WU is a CASP task just by looking at the WU name?
____________

dcdc Profile

Joined: Nov 3 05
Posts: 1596
ID: 8948
Credit: 33,801,861
RAC: 17,327
Message 65887 - Posted 29 Apr 2010 8:22:54 UTC - in response to Message ID 65886.

Just wondering, it is possible tell whether the WU is a CASP task just by looking at the WU name?

Some of the WUs in my queue have CASP in the name now - I don't know if they all will though.
____________

Tom Philippart
Avatar

Joined: May 29 06
Posts: 183
ID: 85247
Credit: 834,667
RAC: 0
Message 65907 - Posted 30 Apr 2010 17:59:50 UTC - in response to Message ID 65811.



We will be utilizing BOINC heavily during the next 4 months, as you can see we already have nearly 4 million workunits lined up.
We have been very busy getting everything ready in time!

Mike


Those 4 million WUs are no CASP WUs???

It says the first targets will be released on may 3rd...
____________

Ed and Harriet Griffith Profile
Avatar

Joined: Sep 17 05
Posts: 39
ID: 82
Credit: 420,649
RAC: 20
Message 65977 - Posted 5 May 2010 13:15:25 UTC

The team BOINC Synergy has selected Rosetta to be its project of the month for May! I hope the extra computing helps with CASP.
____________

David Baker
Forum moderator
Project administrator
Project developer
Project scientist

Joined: Sep 17 05
Posts: 704
ID: 122
Credit: 559,847
RAC: 0
Message 65992 - Posted 6 May 2010 5:57:56 UTC - in response to Message ID 65977.

The team BOINC Synergy has selected Rosetta to be its project of the month for May! I hope the extra computing helps with CASP.



THANKS!!
____________

billy ewell 1931

Joined: Mar 30 07
Posts: 10
ID: 160868
Credit: 3,008,779
RAC: 0
Message 66010 - Posted 6 May 2010 20:38:34 UTC
Last modified: 6 May 2010 20:39:30 UTC

Are we now operating in a primarily CASP9 mode? I have over 220 work tasks queued-up in direct support of R@H and to be crunched by my 20+ cpus but I cannot tell from reviewing the tasks if they are CASP related. I feel very strongly about the efforts of this research program and I intend to fully support the scientific effort with all my resources for the indefinite future.
Thanks to David and Mike and all other operatives who make this project posssible.

Bill

Evan

Joined: Dec 23 05
Posts: 268
ID: 42505
Credit: 402,585
RAC: 0
Message 66011 - Posted 6 May 2010 21:09:35 UTC

Are we now operating in a primarily CASP9 mode
I have recently downloaded a batch with the wording 'casp9' in the description
____________

David Baker
Forum moderator
Project administrator
Project developer
Project scientist

Joined: Sep 17 05
Posts: 704
ID: 122
Credit: 559,847
RAC: 0
Message 66014 - Posted 7 May 2010 3:49:07 UTC

Yes--the first CASP9 target was released a few days ago, and you will be seeing primarily CASP targets for the next month or so. thanks for your contributions to this effort!
____________

Tom Philippart
Avatar

Joined: May 29 06
Posts: 183
ID: 85247
Credit: 834,667
RAC: 0
Message 66030 - Posted 8 May 2010 10:19:43 UTC

Hello, 2 basic questions:

1. are our casp deadlines the "human" or "server" deadlines?

2. where can I find out which target I'm running? I remember on the last CASP challenges, the target name was in cluded in the WU name in this format tXXX_rest_of_the_WU_name. Right now I see a WU with casp9 in the name on my computer, but it says t000...
____________

Mike Tyka

Joined: Oct 20 05
Posts: 96
ID: 5612
Credit: 2,190
RAC: 0
Message 66053 - Posted 10 May 2010 20:10:03 UTC

Hi all!

First of all thanks for all your computing effords!

In answer to some of the questions:

The jobs ending in "casp9" and containing t000_ are homology modelling jobs ones submitted by our automated new rosetta server. Sadly the way it's set up they do not contain the original casp target name but instead are numbered internally so you can't tell from the name what target they are. But you can probably guess from the length and the day they were submitted (the submit day should match up with the target release date).
These targets are operating under the Server deadline, not the human deadline, which means only 3 days from release time to deadline !

OTOH jobs that do not contain casp9 in the name are not necessarily not related to casp. THe human group will be submitting various jobs relating to casp but will use custom names.

Basically a very large fraction of the jobs you'll see on your machines will be relaated to casp in one way or another - i'd say at least 60-70%.


Mike


____________
http://beautifulproteins.blogspot.com/
http://www.miketyka.com/

JChojnacki Profile
Avatar

Joined: Sep 17 05
Posts: 71
ID: 105
Credit: 6,746,778
RAC: 1,867
Message 66056 - Posted 10 May 2010 21:22:18 UTC - in response to Message ID 66053.


Basically a very large fraction of the jobs you'll see on your machines will be relaated to casp in one way or another - i'd say at least 60-70%.

Mike


Thanks for the info, Mike.

____________



Tom Philippart
Avatar

Joined: May 29 06
Posts: 183
ID: 85247
Credit: 834,667
RAC: 0
Message 66067 - Posted 11 May 2010 7:34:14 UTC

hmmm 3 days doesn't seem a lot given the long deadlines of this project.

In this case: @Bakerteam: what's your prefered WU runtime?

I think the goal is to have as many predictions as possible in this time frame...
____________

Murasaki
Avatar

Joined: Apr 20 06
Posts: 303
ID: 78284
Credit: 365,375
RAC: 94
Message 66071 - Posted 11 May 2010 15:16:19 UTC - in response to Message ID 66067.

hmmm 3 days doesn't seem a lot given the long deadlines of this project.


Definitely. I just noticed my slow computer has a "CASP9" task that it has been chugging away on for 4 days. Is the result still valuable? It still has 6 days left before the "official" BOINC deadline.

Would it be worth me cutting the work time in my preferences to force results to come back sooner?
____________

Mike Tyka

Joined: Oct 20 05
Posts: 96
ID: 5612
Credit: 2,190
RAC: 0
Message 66076 - Posted 11 May 2010 22:18:51 UTC

Indeed. For jobs that contain _casp9 in their name, we need the results back within about 18hrs. Everything that comes back after that will ikely not make it into the final analysis - the reason is that we need some time for post analysis here before we can send the final models (we have to choose 5!) to the CASP organisers. So, for those kind of WUs it would help us tremendously if you could set your runtimes to 12-16hrs, even if you only end up crunching a single decoy (sometimes we get fairly large structures from CASP).
These deadline restrictions are only for the server predictions - the human team have about 3 weeks per target, but they use a different naming scheme.

Mike


____________
http://beautifulproteins.blogspot.com/
http://www.miketyka.com/

Sid Celery

Joined: Feb 11 08
Posts: 796
ID: 241409
Credit: 9,545,746
RAC: 7,447
Message 66078 - Posted 12 May 2010 2:45:13 UTC - in response to Message ID 66076.

Indeed. For jobs that contain _casp9 in their name, we need the results back within about 18hrs. Everything that comes back after that will ikely not make it into the final analysis - the reason is that we need some time for post analysis here before we can send the final models (we have to choose 5!) to the CASP organisers. So, for those kind of WUs it would help us tremendously if you could set your runtimes to 12-16hrs, even if you only end up crunching a single decoy (sometimes we get fairly large structures from CASP).
These deadline restrictions are only for the server predictions - the human team have about 3 weeks per target, but they use a different naming scheme.

Mike, unless I'm misunderstanding something here, doesn't this also mean that the work buffer needs to be adjusted too? If, for example, I have a 2.5 day work buffer, it'll be about that amount of time before a WU gets to the top of the pile to even start running. Then, if I have an 8-hour runtime, the completed WU won't be ready to return until 2.83 days after being received.

If I understand you right, 18 hours (=0.75 days) is the maximum that will fall within this window. So anything more than a 0.5 day buffer and 6-hour runtime (or some combination > 0.75days\18 hours) is guaranteed to be late.

An early reply will be appreciated. I'll be out of town in 12 hours until late Sunday and won't have an opportunity to change my local settings between those dates. It sounds like everything I'm crunching will be of no use at all. Please tell me I'm wrong!
____________

Mod.Sense
Forum moderator
Project administrator

Joined: Aug 22 06
Posts: 3381
ID: 106194
Credit: 0
RAC: 0
Message 66082 - Posted 12 May 2010 3:53:13 UTC

Certainly correct, good catch Sid. If all of the tasks you download are CASP9 tasks, then a large download buffer defeats the purpose of the shorter deadline.
____________
Rosetta Moderator: Mod.Sense

Venturini Dario[VENETO] Profile

Joined: May 25 07
Posts: 22
ID: 179805
Credit: 245,028
RAC: 0
Message 66086 - Posted 12 May 2010 10:10:36 UTC - in response to Message ID 66082.

Certainly correct, good catch Sid. If all of the tasks you download are CASP9 tasks, then a large download buffer defeats the purpose of the shorter deadline.


So, do we really need to set cache at 0, connection "always on" and "target run time" at 12 hours, else our WUs will be useless?

Murasaki
Avatar

Joined: Apr 20 06
Posts: 303
ID: 78284
Credit: 365,375
RAC: 94
Message 66087 - Posted 12 May 2010 10:15:14 UTC - in response to Message ID 66078.

It sounds like everything I'm crunching will be of no use at all. Please tell me I'm wrong!


You are wrong. ^_^

I took a look at the tasks you received. Only 1 has "casp9" in the description and will be restricted to the 18 hour window. From what Mike has said the rest should be under the human prediction deadline and have a few weeks to get the results back.
____________

Max DesGeorges

Joined: Oct 1 05
Posts: 35
ID: 2201
Credit: 942,527
RAC: 0
Message 66088 - Posted 12 May 2010 12:10:13 UTC - in response to Message ID 66087.

Mike, if you need the casp9 WUs to be finished within a day, why you don't set for this WUs a different deadline (one day or less if possible) and send them only to users that give back results rapidly?
____________

billy ewell 1931

Joined: Mar 30 07
Posts: 10
ID: 160868
Credit: 3,008,779
RAC: 0
Message 66097 - Posted 12 May 2010 19:22:00 UTC

Based on my reading and interpretation of the preceding posts, I am resetting my account "Computing Preferences" to "0" for the option that says: "Maintain enough work for an additional _______days". My previous setting was 1.5 days so to hopefully guarantee my computers would not be deprived of work during server down-times; and in so doing I was unknowingly guaranteeing that many of my random casp9 units would exceed the 18-hour time restraint from download to upload.

Secondly; wouldn't it be practical for admin to post clear casp9 processing time limitations to the Home news section.

I appreciate the feedback from the previous posters.

Sid Celery

Joined: Feb 11 08
Posts: 796
ID: 241409
Credit: 9,545,746
RAC: 7,447
Message 66098 - Posted 12 May 2010 22:37:03 UTC - in response to Message ID 66097.
Last modified: 12 May 2010 22:38:50 UTC

Certainly correct, good catch Sid. If all of the tasks you download are CASP9 tasks, then a large download buffer defeats the purpose of the shorter deadline.

Nice that I understood it. Not nice that I left home before checking what the answer was :( I can adjust the dual-core laptop I'm using now but that's just 1\3 of my processing.

You are wrong. ^_^

I took a look at the tasks you received. Only 1 has "casp9" in the description and will be restricted to the 18 hour window. From what Mike has said the rest should be under the human prediction deadline and have a few weeks to get the results back.

:) Thanks for checking for me. I wasn't sure what to look at - I wasn't following the discussion above very well. Fortunately (or not) BOINC was going through another phase of not grabbing WUs when you looked at my quad desktop. Now there are over 20 8-hour WUs (still only 1 CASP9) - I'm just lucky that's the case.

Mike, if you need the casp9 WUs to be finished within a day, why you don't set for this WUs a different deadline (one day or less if possible) and send them only to users that give back results rapidly?

This makes sense. People can't be expected to micro-manage their tasks, nor to identify which task is of which type - this is best set by the project when a certain batch of tasks are released. Changing the work-buffer is a very blunt instrument, as is changing the runtime.

Based on my reading and interpretation of the preceding posts, I am resetting my account "Computing Preferences" to "0" for the option that says: "Maintain enough work for an additional _______days". My previous setting was 1.5 days so to hopefully guarantee my computers would not be deprived of work during server down-times; and in so doing I was unknowingly guaranteeing that many of my random casp9 units would exceed the 18-hour time restraint from download to upload.

On the plus side, BOINC defaults are for a 0.25 day buffer and a 3-hour runtime, so the majority of users won't have a problem. And if there are a whole bunch of tasks available (which there are) we'll only run dry if the server goes down.

The problem is for all us inveterate tinkerers. I'm cutting back to 0.33 days buffer to go with my 8 hour runtimes (in case of over-runs and to give time to get the WUs back to base).
____________

billy ewell 1931

Joined: Mar 30 07
Posts: 10
ID: 160868
Credit: 3,008,779
RAC: 0
Message 66100 - Posted 13 May 2010 2:25:56 UTC

That is one hell of a fine post Sid; great job.

Murasaki
Avatar

Joined: Apr 20 06
Posts: 303
ID: 78284
Credit: 365,375
RAC: 94
Message 66111 - Posted 13 May 2010 22:50:37 UTC

I just spotted that the graphics of a casp 9 task I have is showing both energy levels and RMSD. As CASP9 targets are unknown I had thought the RMSD comparison was impossible.

Is this really a CASP9 target or just a non-casp target with a poor choice of name?

If it is CASP have you guys come up with some idea for estimating the RMSD and trying to track against it?
____________

l_mckeon

Joined: Jun 5 07
Posts: 44
ID: 182403
Credit: 180,717
RAC: 0
Message 66430 - Posted 3 Jun 2010 1:05:18 UTC - in response to Message ID 66111.

I just spotted that the graphics of a casp 9 task I have is showing both energy levels and RMSD. As CASP9 targets are unknown I had thought the RMSD comparison was impossible.


Ditto, I've been wondering that as well.

Aren't CASP targets supposed to be blind?

Mod.Sense
Forum moderator
Project administrator

Joined: Aug 22 06
Posts: 3381
ID: 106194
Credit: 0
RAC: 0
Message 66433 - Posted 3 Jun 2010 3:50:37 UTC

CASP, while in progress, issues blind targets. Or targets that noone on earth knows the structure for. If someone solves the structure during the period of time when submissions can still be made to CASP then the protein is thrown out (sometimes scientists elsewhere in the world happen to be working on solving the same protein that CASP is following and finish before the folks that CASP is working with).

Sometimes targets from prior CASP exercises are studied to see if more insight can be gleaned. These might say casp8 in the name.

I am most accustomed to tasks with no known solutions not displaying an RMSD at all. But from Mike's post here it sounds like some tasks show a flat line for RMSD instead when the target's structure is unknown.
____________
Rosetta Moderator: Mod.Sense

Murasaki
Avatar

Joined: Apr 20 06
Posts: 303
ID: 78284
Credit: 365,375
RAC: 94
Message 66446 - Posted 3 Jun 2010 19:53:19 UTC - in response to Message ID 66433.

I am most accustomed to tasks with no known solutions not displaying an RMSD at all. But from Mike's post here it sounds like some tasks show a flat line for RMSD instead when the target's structure is unknown.


No, the ones I was watching at the time of my last post definitely had known structure with a "native" graphic and changing RMSD. However, a few days later I stopped getting casp9s with RMSD and instead started picking up tests using casp8 models.

I suspect what I saw was just some casp8 targets with a typo in the batch name.
____________

cenit Profile

Joined: Apr 1 07
Posts: 13
ID: 161706
Credit: 1,630,287
RAC: 0
Message 66557 - Posted 12 Jun 2010 10:28:30 UTC - in response to Message ID 65992.
Last modified: 12 Jun 2010 10:28:46 UTC

The team BOINC Synergy has selected Rosetta to be its project of the month for May! I hope the extra computing helps with CASP.

our team (BOINC.Italy) yesterday was the most productive team worldwide relating to BOINC credits...
wooo!
let's help Baker team and their great research effort in this CASP challenge!

transient
Avatar

Joined: Sep 30 06
Posts: 376
ID: 115553
Credit: 7,590,569
RAC: 2,277
Message 66839 - Posted 10 Jul 2010 14:58:58 UTC

How are things going on the CASPP9 front? We're not seeing a lot of information on that subject lately.
____________

Jeff Gilchrist

Joined: Oct 7 05
Posts: 33
ID: 3191
Credit: 2,398,990
RAC: 0
Message 66857 - Posted 12 Jul 2010 13:52:56 UTC

Does anyone know how long Rosetta@Home will be handing out CASP9 WUs? Will they be coming out until the end of July, mid-August, end of August?

Thanks.

____________

Murasaki
Avatar

Joined: Apr 20 06
Posts: 303
ID: 78284
Credit: 365,375
RAC: 94
Message 66865 - Posted 12 Jul 2010 23:19:33 UTC - in response to Message ID 66857.

Does anyone know how long Rosetta@Home will be handing out CASP9 WUs? Will they be coming out until the end of July, mid-August, end of August?


From the first post, CASP9 started around the beginning of May and the team will be sending work out through BOINC for 4 months. That would suggest the CASP work should be tailing off towards the end of August/start of September.
____________

Sid Celery

Joined: Feb 11 08
Posts: 796
ID: 241409
Credit: 9,545,746
RAC: 7,447
Message 67157 - Posted 10 Aug 2010 11:42:22 UTC - in response to Message ID 66865.

Does anyone know how long Rosetta@Home will be handing out CASP9 WUs? Will they be coming out until the end of July, mid-August, end of August?

From the first post, CASP9 started around the beginning of May and the team will be sending work out through BOINC for 4 months. That would suggest the CASP work should be tailing off towards the end of August/start of September.

I'm guessing this is why we're now seeing the queue of jobs dropping from 4m+ to 3m.

Can we have a statement of where we stand on CASP9 jobs now? Are the deadlines still at 18 hours? Are we keeping up with the schedule of remaining jobs?

And after it's all over, will the project look to maintain a large queue for ongoing work? It seems to have been very successful these last few months.
____________

Murasaki
Avatar

Joined: Apr 20 06
Posts: 303
ID: 78284
Credit: 365,375
RAC: 94
Message 67159 - Posted 10 Aug 2010 19:40:22 UTC

It looks like the deadline for submitting CASP9 results was the first week of August. I assume that we are now back to normal crunching operations, but it would be nice if we could get confirmation from the project team.
____________

Nickhorsky

Joined: Mar 11 07
Posts: 10
ID: 152071
Credit: 134,214
RAC: 0
Message 67166 - Posted 12 Aug 2010 3:20:57 UTC

0
Server groups: CLOSED on July 24, 2010

Human groups: OPEN until August 20, 2010

Server groups: CLOSED on July 24, 2010

Human groups: CLOSED for regular predictions on July 31, 2010. OPEN until
August 20, 2010 for the refinement experiment.

Will start collecting on August 30, 2010.

Deadline for submission - September 20, 2010.

Registration will open on September 21, 2010
Financial support application deadline - October 6.
Notification of the fellowship recipients - October 15.
Early registration deadline - October 22.
Meeting - December 5-9, 2010, Pacific Grove, California, USA.

Mark

Joined: Nov 10 08
Posts: 2
ID: 287398
Credit: 1,108,348
RAC: 0
Message 67522 - Posted 1 Sep 2010 15:55:47 UTC - in response to Message ID 67166.
Last modified: 1 Sep 2010 15:56:18 UTC

Do we have any results on how well rosetta did compared to UW's human team or compared to Foldit, or other research teams?

transient
Avatar

Joined: Sep 30 06
Posts: 376
ID: 115553
Credit: 7,590,569
RAC: 2,277
Message 67523 - Posted 1 Sep 2010 16:36:03 UTC
Last modified: 1 Sep 2010 16:38:21 UTC

I imagine intermediate results of CASP9 are available. But as far as I know the definitive report is scheduled for December. Go to this site for more info.
http://predictioncenter.org/casp9/
____________

David Baker
Forum moderator
Project administrator
Project developer
Project scientist

Joined: Sep 17 05
Posts: 704
ID: 122
Credit: 559,847
RAC: 0
Message 67605 - Posted 6 Sep 2010 6:43:15 UTC

The results are trickling in and looking pretty good, but the definitive analysis won't be available until December at the CASP meeting
____________

P . P . L .
Avatar

Joined: Aug 20 06
Posts: 581
ID: 105843
Credit: 4,864,105
RAC: 0
Message 67669 - Posted 9 Sep 2010 21:58:13 UTC

Hi.

I'm seeing more tasks with casp9 marked in them now, then when CASP was running how come.

____________


Mod.Sense
Forum moderator
Project administrator

Joined: Aug 22 06
Posts: 3381
ID: 106194
Credit: 0
RAC: 0
Message 67690 - Posted 10 Sep 2010 17:50:06 UTC

It is always good to study further things that have challenged you in the past. At least that was the reason after CASP8. They were seeing if new tweaks would improve the prediction and how much the prediction would have improved with existing methods if there were more crunching time before submitting their prediction.
____________
Rosetta Moderator: Mod.Sense

transient
Avatar

Joined: Sep 30 06
Posts: 376
ID: 115553
Credit: 7,590,569
RAC: 2,277
Message 68841 - Posted 19 Dec 2010 9:29:22 UTC - in response to Message ID 67605.

The results are trickling in and looking pretty good, but the definitive analysis won't be available until December at the CASP meeting


As far as I can tell, the results are in. The meeting has been held. How was the meeting? Did Rosetta perform as expected? What have you learned?

____________

Polian
Avatar

Joined: Sep 21 05
Posts: 152
ID: 702
Credit: 9,832,077
RAC: 1,064
Message 69092 - Posted 9 Jan 2011 19:05:21 UTC

I'm curious as well.
____________

David Baker
Forum moderator
Project administrator
Project developer
Project scientist

Joined: Sep 17 05
Posts: 704
ID: 122
Credit: 559,847
RAC: 0
Message 69175 - Posted 11 Jan 2011 6:36:39 UTC

The CASP9 meeting was very interesting.

The general consensus of the assessors this year was that there were a number of groups "at the top", but there were no outright winners-there is a bit of bunching up and many groups did very well by successfully picking from among the models submitted by automated servers (you can imagine this led to a lot of discussion at the meeting). While no group clearly outperformed other groups by a significant margin, the best ab initio structure prediction at the meeting stood out quite clearly. This was a Rosetta@home product entirely-the rosettaServer this year sent jobs out to rosetta@home and then submitted models from the lowest energy clusters. this was without a doubt the best automated ab initio structure prediction in casp.

In contrast to our approach with rosetta@home which carries out a large scale search for the lowest energy structure, other groups have developed methods that cleverly exploit information present in the protein structure database, and with the increase in the size of this database, these methods keep working better. I am now very interested in using such information to help guide the search for the lowest energy structure (which is very hard for large proteins), in the same way we have been using experimental data to guide search. Some of the jobs going out in the next months on rosetta@home will be testing approaches for taking advantage of this information.

I apologize for the recent system crash. we are beefing up our servers considerably so hopefully this won't happen again for a long time to come.
____________

rochester new york Profile
Avatar

Joined: Jul 2 06
Posts: 2562
ID: 98229
Credit: 958,139
RAC: 127
Message 69435 - Posted 20 Jan 2011 6:35:32 UTC

http://www.redorbit.com/news/health/1975337/university_of_michigan_lab_takes_top_honors_in_worldwide_protein/index.html?source=r_health

Ed and Harriet Griffith Profile
Avatar

Joined: Sep 17 05
Posts: 39
ID: 82
Credit: 420,649
RAC: 20
Message 69440 - Posted 20 Jan 2011 15:16:31 UTC

Thanks Rochester for a very interesting article on what the University of Michigan is accomplishing in protein folding!
____________

[VENETO] boboviz Profile

Joined: Dec 1 05
Posts: 545
ID: 25524
Credit: 1,510,213
RAC: 1,277
Message 69760 - Posted 8 Mar 2011 19:02:39 UTC

I downoad today some _casp9_ wu. Is not finished casp9 competition??
____________

Murasaki
Avatar

Joined: Apr 20 06
Posts: 303
ID: 78284
Credit: 365,375
RAC: 94
Message 69761 - Posted 8 Mar 2011 19:42:06 UTC - in response to Message ID 69760.

I downoad today some _casp9_ wu. Is not finished casp9 competition??


Yes, CASP 9 is finished, but they often repeat old CASP tasks to see if their new ideas would have produced better results.

[VENETO] boboviz Profile

Joined: Dec 1 05
Posts: 545
ID: 25524
Credit: 1,510,213
RAC: 1,277
Message 75624 - Posted 19 May 2013 6:21:44 UTC - in response to Message ID 69761.

Yes, CASP 9 is finished, but they often repeat old CASP tasks to see if their new ideas would have produced better results.


Still CASP9 wus......after 3 yesars!!
____________

krypton
Forum moderator
Project administrator
Project developer
Project scientist

Joined: Nov 16 11
Posts: 105
ID: 436004
Credit: 1,813,153
RAC: 2,013
Message 75631 - Posted 20 May 2013 6:44:10 UTC - in response to Message ID 75624.

Yes, CASP 9 is finished, but they often repeat old CASP tasks to see if their new ideas would have produced better results.


Still CASP9 wus......after 3 yesars!!


You can never get enough benchmark sets, its surprising how different protein target are from one CASP to another. I personally pull targets from CASP9/CASP10 and CAMEO (another competition we take part in), to get enough variation.

On a general note, we can get the computer to say almost anything we want it to say, we can adjust our parameters endlessly until we get exactly what we want... but this is called "over-fitting." And parameters calculated for one set might not work for another or a completely new target. The purpose of CASP was to provide a blind test to test our protocols.

But during development, as we wait for the next CASP, we once again risk over-fitting, especially if we use only a single dataset. We thus put aside certain targets (CASP9 etc.) adjust our parameters to work well on one set and test it independently on another set.

So every time you see something running with "CASP9" in the title that means we are testing something new!

Message boards : Rosetta@home Science : CASP9


Home | Join | About | Participants | Community | Statistics

Copyright © 2017 University of Washington

Last Modified: 10 Nov 2010 1:51:38 UTC
Back to top ^