Wiggle lock on 499 (again)

Case number:845813-991397
Topic:Game: Tools
Opened by:Rav3n_pl
Status:Closed
Type:Bug
Opened on:Saturday, December 24, 2011 - 01:43
Last modified:Tuesday, May 15, 2012 - 22:32

Wiggle is "locked" on middle of puzzle.
If we rebuild middle segments thingy is dancing around them when wiggled.

(Sat, 12/24/2011 - 01:43  |  10 comments)


beta_helix's picture
User offline. Last seen 2 days 8 hours ago. Offline
Joined: 05/09/2008
Groups: None
Status: Open » Closed

we were unable to reproduce this bug.

Let us know if it happens on the similar puzzle 503: http://fold.it/portal/node/991538

Joined: 06/17/2010
Status: Closed » Open

Yap, 503 have same bug.
Just reset, wiggle and watch. 16/17 is locked in space.
Wiggling: http://youtu.be/ohDt_weM36c
After rebuild: http://youtu.be/tYj2NOqyVCM
After hand pulls: http://youtu.be/7LJKy9pT914

Dev client.

beta_helix's picture
User offline. Last seen 2 days 8 hours ago. Offline
Joined: 05/09/2008
Groups: None
Status: Open » Open

does that happen in main?
re-assigning...

Joined: 06/17/2010

Yes, it looks exactly the same in main.

Joined: 06/17/2010

Bump. It can be observed on every puzzle.
Yes, we can try to accept is as a feature, but I think if we move this "null point" to another location (ie best scoring segment) wiggle can work much better, especially when we are rebuilding this "locked" part.
If it is possible please include it to dev preview for testing when we will have some small one like R003 (up to 100 residues anyway).

jflat06's picture
User offline. Last seen 2 days 8 hours ago. Offline
Joined: 09/29/2010
Groups: Window Group

The location of where this "lock" occurs only matters in situations where the absolute position of the protein in space matters.

That is to say, symmetry, electron density, puzzles with multiple chains (and possibly some others).

In those puzzles, no segment should be "locked in space", since the entire protein is allowed to move during wiggle.

For all other puzzles, the position of where this segment is within the protein has no effect on the minimization; it is purely chosen based on cosmetic considerations. The minimizer is simply minimizing the torsion angles of the residues, and the position of the root residue has no effect on this.

Joined: 06/17/2010

Well if it is only cosmetic and can be set by puzzle configuration can you please post 2 identical small (around 50 residues) puzzles for dev? One "normal" hooked in the middle and second hooked somewhere else?
Also allow (if it will be possible) to load solutions between that 2 puzzles or make "broken in the middle" start configuration.
Just to confirm that it really has nothing to score.
Also in some cases (rebuild middle) locking it somewhere else will prevent twisting and dancing when wiggle.

spmm's picture
User offline. Last seen 36 weeks 2 days ago. Offline
Joined: 08/05/2010
Groups: Void Crushers

same on 528 all templates

alwen's picture
User offline. Last seen 6 weeks 5 days ago. Offline
Joined: 10/03/2011

I'm seeing a wiggle lock on 548 (segment 17).

I'm on Win 7 start in main.

Joined: 06/17/2010
Status: Open » Closed

Duplicate to:
http://fold.it/portal/node/992677

Sitemap

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, Amazon, Microsoft, Adobe, RosettaCommons