573b solutions load with two bad spots in 580

Case number:671071-992994
Topic:Game: Other
Opened by:Bletchley Park
Opened on:Sunday, June 17, 2012 - 09:13
Last modified:Monday, June 18, 2012 - 18:17

The solutions for 573b are saved without problem and in good shape.
When loaded into 580 there are two distinct areas that score very badly, -200.000 points.

This seems like a bug to me.

(Sun, 06/17/2012 - 09:13  |  7 comments)

Joined: 05/19/2009
Groups: Contenders

residues 88 (-26238),89 (-26216) and 129 (-81012), 130 (-81036)

Joined: 05/19/2009
Groups: Contenders

More info: the difference in scoring between 573b and 580 is completely in the backbone score.
I loaded the solution in 2 clients, one running 573b the other running 580 and compared the scores.

I also noticed this:
Residu 129 does not have a backbone score in 573b... but has a negative score in 580.

Neither do 1,5,11,17,19, 20, 32 ... 74,75,79 in 573b. I stopped checking.
Surely this must be a display issue with the Tab window ?

brow42's picture
User offline. Last seen 2 years 1 week ago. Offline
Joined: 09/19/2011
Groups: None

The tab window does not show a score if it is exactly 0 (after rounding to integer). A little annoying since it makes scores move around in the window. So the question is, are these score ~0, and why.

Mark-'s picture
User offline. Last seen 21 weeks 2 days ago. Offline
Joined: 08/05/2009
Groups: Contenders

Just to add to this.

The saved solution was from my 573b solo in which I had manually added cutpoints at those segments.

Loading my best solo from 573b in 580 shows a score of -200K points with similar bad areas where I had previously used cutpoints. This appears to be an artifact of using cutpoints.

beta_helix's picture
User offline. Last seen 4 days 1 hour ago. Offline
Joined: 05/09/2008
Groups: None

hopefully we can figure out this bug ASAP!

Joined: 09/21/2011
Groups: Void Crushers

I had the same experience, but after fixing the bad spots I was able to get the puzzle about 50 points higher then I did in 573b. This makes me wondering if this effect can be used to make a new function to reinit the backbone while keeping the main structure in place? (Re-align to itself?)

jflat06's picture
User offline. Last seen 1 day 2 hours ago. Offline
Joined: 09/29/2010
Groups: Window Group

This appears to be caused by differences in the puzzle setup.

The recommended workaround is to cut the portions of the protein that are scoring poorly, wiggle, and then rejoin them.


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, Boehringer Ingelheim, RosettaCommons