MantisBT - Rosetta
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0000255Rosetta[All Projects] Incorrect Resultspublic2013-04-25 19:182013-04-28 17:40
Reporteramelie 
Assigned Toamelie 
PrioritynormalSeverityminorReproducibilityalways
StatusresolvedResolutionfixed 
PlatformAll platformsOSAnyOS VersionAny
Product VersionTrunk 
Fixed in Version 
Application(s) Affectedloopmodel.<binary>
Command Line Usedloopmodel.<binary> -loops:remodel perturb_kic -loops:refine refine_kic
Developer OptionsConfirmed As Bug
Fixed in SVN Version55062
Summary0000255: KIC always reports a chainbreak score of 0
DescriptionThe chainbreak score term of the scorefxn used for repacking must have gotten lost in some repacking step, so that now the reported chainbreak score is always 0.
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2013-04-25 19:18amelieNew Issue
2013-04-25 19:18amelieStatusnew => assigned
2013-04-25 19:18amelieAssigned To => amelie
2013-04-25 21:05smlewisNote Added: 0000260
2013-04-25 21:14amelieNote Added: 0000261
2013-04-28 17:33amelieNote Added: 0000263
2013-04-28 17:34amelieFixed in SVN Version => 55062
2013-04-28 17:34amelieStatusassigned => closed
2013-04-28 17:34amelieResolutionopen => fixed
2013-04-28 17:39amelieStatusclosed => feedback
2013-04-28 17:39amelieResolutionfixed => reopened
2013-04-28 17:40amelieStatusfeedback => resolved
2013-04-28 17:40amelieResolutionreopened => fixed

Notes
(0000260)
smlewis   
2013-04-25 21:05   
If the chainbreak score is reported, the score is working - the problem is the protein_cutpoint_upper and _lower variant types on the expected break are lost. chainbreak is only calculated over peptide bonds with those variants (and a break in the FoldTree/AtomTree to go with it, or else the variants cannot be added - since KIC has no break, it might never work).
(0000261)
amelie   
2013-04-25 21:14   
Steven - this was probably a misunderstanding, the chainbreak score is not reported during the course of loop modeling, only at the very end of the protocol where it is explicitly requested via getPoseExtraScores. When using sfxn->show(pose) there is no chainbreak score shown. The upper and lower cutpoints are set though. As soon as I set a non-zero weight for chainbreak, there is a non-zero score.
(0000263)
amelie   
2013-04-28 17:33   
resolved in r55062