Wiggle can decrease score when there are filters

Case number:845813-994912
Topic:Game: Tools
Opened by:wozzarelli
Status:Closed
Type:Bug
Opened on:Saturday, April 6, 2013 - 13:18
Last modified:Monday, April 8, 2013 - 01:34

Showed on Mac version (up to date as of 6/4/13).
Occurred in Puzzle 691 - 60 Residue Tetramer

My overall score decreased by nearly 10 points after performing a wiggle, due to the combined effect of two penalties. There were no constraints imposed, ie. no bands or frozen segments, slow filters were enabled and clashing importance was equal to 1.

Arithmetic shows that the score before penalties would have increased slightly, so I assume this is the criteria for a "successful" wiggle. I suggest wiggle takes into account the score after penalties, or that it's optional.

I didn't attach a screenshot- the only thing that looked different was the score.

Arithmetic:

BEFORE WIGGLE: 12877.108 (would have been 13687.108 without penalties)
Score:
0 Bonuses, 3 penalties, all conditions met
- Residue IE Score: met
- Core Existence; lost (-360)
- Polar Core: lost (-450)

AFTER WIGGLE: 12867.153 (would have been 13687.153 without penalties)
Score:
0 Bonuses, 3 penalties, all conditions met
- Residue IE Score: met
- Core Existence; lost (-320)
- Polar Core: lost (-500)

(Sat, 04/06/2013 - 13:18  |  3 comments)


Joined: 09/21/2011
Groups: Void Crushers

Wiggle does micro optimalisations. But what is good for the segments involved can have a negative effect on the filters. Wiggle does not care. It is a feature. It would probably make wiggle a lot slower if it has to take the filters also in account.

Joined: 05/27/2010
Groups: None

OK

Joined: 04/19/2009
Status: Open » Closed

Good.

Please close your cases when your questions are answered :-)

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