MantisBT - Rosetta
View Issue Details
0000029Rosetta[All Projects] Crashpublic2012-06-26 13:172013-05-21 09:51
lowcrashunable to reproduce
pmut_scan_parallel, ??
0000029: "Problem in calculating overlap between atoms. Termination calculation."
(from the boards, [^])

Hi all,
I'm trying to run pmut_scan_parallel application with "-score:weights design_hpatch.wts" scoring function, as suggested in manual. However, it always exits from: src/core/scoring/ line 215 and an error reported:
"Problem in calculating overlap between atoms. Termination calculation.
radius_a: 3.16, radius_b: 3.16, distance_ijxyz: 14.199, cosine theta: 2.245"
Does it mean that atom a and b are too far from each other so that there is no overlap between them? Why it happens in hpatch scoring, and any suggestion to avoid it?
Many thanks,
Ron says he knows how to fix it and will when he gets a chance
No tags attached.
Issue History
2012-06-26 13:17smlewisNew Issue
2012-06-26 13:17smlewisStatusnew => assigned
2012-06-26 13:17smlewisAssigned To => ronj
2013-02-05 17:24ronjStatusassigned => confirmed
2013-02-05 17:26ronjFixed in SVN Version => 53570
2013-02-05 17:26ronjNote Added: 0000139
2013-02-05 17:27ronjStatusconfirmed => closed
2013-02-05 17:27ronjResolutionopen => fixed
2013-02-24 20:38smlewisNote Added: 0000142
2013-02-24 20:38smlewisStatusclosed => resolved

2013-02-05 17:26   
Problem was with re-initialization of certain variables in the HPatchIG when multiple packing cycles (ndruns, not common) or the MultiCoolAnnealer are in use. Rosetta indicates the bug occurs when it's calculating atom-atom overlaps for SASA determination, even though it has nothing to do with that. Fixed by re-initing the right vectors in the HPatchIG after each annealing run.
2013-02-24 20:38   
change closed -> resolved (was fixed, not a nonbug)