puzzle picture
897: Large CASP11 Refinement Puzzle: TR760
Status: Closed


Name: 897: Large CASP11 Refinement Puzzle: TR760
Status: Closed
Created: 05/15/2014
Points: 100
Expired: 05/30/2014 - 10:00
Difficulty: Advanced
Description: The first CASP11 refinement target is larger than most Foldit puzzles, at 201 residues. This model was the best server prediction for target T0760, generated by the MULTICOM-NOVEL server. We've been told that ~58% of this model matches the native! As this is such a large protein, we are giving you 10 days to work on it. More info in the puzzle comments, where you can post your feedback about whether puzzles of this size are too much for us to tackle in CASP11.
Categories: CASP11, Overall, Prediction

Top Groups

1Anthropic Dreams10,878100
3Void Crushers10,73670
5Beta Folders10,71447

Top Evolvers

Top Soloists

Need this puzzle? Log in to download.  


beta_helix's picture
User offline. Last seen 5 days 11 hours ago. Offline
Joined: 05/09/2008
Groups: None
More info here:

This is the page for this CASP11 refinement target:

Note that the organizers gave us this hint:

"Starting model's GDT_HA=58"

GDT_TS=58 essentially means that 58% of the model superimposes correctly onto the native structure: http://en.wikipedia.org/wiki/Global_distance_test

beta_helix's picture
User offline. Last seen 5 days 11 hours ago. Offline
Joined: 05/09/2008
Groups: None
We haven't posted such a big puzzle since the NewChapter update

So please give us your feedback about whether or not 10 days is sufficient for such a large puzzle, or if you feel we shouldn't participate at all in these larger CASP11 refinement puzzles.

For example, the second refinement target, TR762, has 257 residues... so we will NOT be posting that one!

Thank you for your input...

smilingone's picture
User offline. Last seen 2 hours 47 min ago. Offline
Joined: 09/19/2011
Groups: Beta Folders
Five days/ten days...that

Five days/ten days...that isn't the issue. This size is simply too large to run with other puzzles. Since you asked, my vote is no to this size.

Joined: 09/21/2011
just to clarify.

58 percent would most likey be similar to tracing a line and staying on the line 58 percent of the time? not so much that large sections are spot on, and the there is a forty two percent section that needs to be worked on? i only ask because segment 75 is looking like a nice place for a cutpoint and then adjusting both halves of the protein to mesh better with each other.

Joined: 09/21/2011
after wiggling it out, i see

after wiggling it out, i see that it wouldn't be wise not to make major renovations. refinement of the typical kind is most likely.

bkoep's picture
User offline. Last seen 1 day 9 hours ago. Offline
Joined: 11/15/2012
Groups: None
Secondary structure predictions

Here is the sequence logo predicted by the SAM server:

H = helix
E = sheet
C = loop (or coil)

The taller the letter at each position, the higher the probability of that specific secondary structure for that amino acid.

Note that these secondary structure predictions are not necessarily present in the starting structure that has been provided in the puzzle.


If you are a new Foldit player, please make sure you have completed all these tutorials before trying this puzzle:
- lonely sheets,
- lock and lower,
- rebuild,
- structure and rebuild,
- control over clashing.

Joined: 09/24/2012
Groups: Go Science

Thanks bkoep for the secondary structure reduction.
May I suggest you to provide also the sequence in writing, so that we can immediately copy it into our puzzle using available scripts?

beta_helix's picture
User offline. Last seen 5 days 11 hours ago. Offline
Joined: 05/09/2008
Groups: None
You can find the sequence here, Bruno:

Click on the CASP link on the first comment:

and it will point you to this page with the sequence:

Joined: 09/24/2012
Groups: Go Science
secoundary structure

Thnks, but I thought te secondary structure in writing ...
This would be easy to "copy-paste" for such a big puzzle.

Joined: 01/23/2014
Groups: None
Cut and Create

Could you not create a tool that allows cut sections of the protein to be saved as new tracks ? ie

Cut at residue 7
Cut at residue 54
Right click context menu
Save cut section as...track 1


frood66's picture
User offline. Last seen 11 hours 50 min ago. Offline
Joined: 09/20/2011
Groups: Marvin's bunch
Puzzles of this size were

Puzzles of this size were pretty well unplayable in Legacy. NC only makes the situation substantially worse. Even puzzle 894 at 163 segs is now 'pushing it'. Playable only to a moderate level of investigation by many - and this at the cost of running just one client and ignoring other puzzles.

Joined: 06/06/2013
Groups: Gargleblasters
Puzzle too large for older machines

I keep crashing, even with only 1 client running a single program. This is frustrating. At this point, I'd love to get zeroed out of this puzzle altogether
It would be interesting to be able to freeze and possibly work the puzzle in pieces before putting it all together. but after tonight, I may just abandon this as a bad match for my equipment and do something that I can do on my ancient pc

Joined: 06/06/2013
Groups: Gargleblasters
surrender on 897

puzzle too big, even if I freeze half and work on half. quitting in disappointment. can't work it
I wanted to get a structure in place and go with it. perhaps in the future either the NC or my machine will make something like this work :-(

Joined: 03/28/2012
Groups: Go Science

it actually keeps crashing and closing its self for me. ill run multiple clients or one client and this is the only puzzle that crashes

Joined: 03/28/2012
Groups: Go Science

im running a new machine (quad core, 8 threads, 8 gb of ram) so....
(sorry for the double post)

beta_helix's picture
User offline. Last seen 5 days 11 hours ago. Offline
Joined: 05/09/2008
Groups: None
Hopefully this helps:

If you are having trouble with crashes on this puzzle, NickyCGS has found a temporary fix that could help you:


Get Started: Download
  Windows    OSX    Linux  
(10.7 or later)

Are you new to Foldit? Click here.

Are you a student? Click here.

Are you an educator? Click here.
Only search fold.it
Recommend Foldit
User login
Top New Users

Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Microsoft, Adobe, RosettaCommons