[gmx-users] 3.3.2 mdrun problems

[Mitchell Stanton-Cook]

Dear all,


I have been running GMX331. Recently I have encountered a documented 
restraint problem: "More than 8 graph edges per atom". See:

* http://www.gromacs.org/pipermail/gmx-developers/2007-March/002060.html
* http://www.gromacs.org/pipermail/gmx-developers/2006-April/001591.html

Rather than modify the code as suggested in those threads (as I was 
unsure of the results) my first thought was to use the newly released 
GMX3.3.2 as I assumed that the problem had been submitted to bugzilla 
and thus fixed in 3.3.2.

However, my problems have worsened - mdrun in 3.3.2 appears to have a 
problem. This is the output after beginning a production run:


 5000000 steps,  10000.0 ps.


 -------------------------------------------------------
 Program mdrun, VERSION 3.3.2
 Source code file: network.c, line: 437

 Routine should not have been called:
 gmx_sumi


I am running ordinary solvated protein molecular dynamics using both 
distance and orientation restraints.


My control simulation - which does not have the "More than 8 graph edges 
per atom" problem and therefore runs fine in 3.3.1 also flags the same 
error in mdrun 3.3.2.


I have also installed GMX3.3.2 on 3 different machines and still have 
the same problems.


I have searched google but no hits yet.

Anyone with ideas?


Cheers Mitch


-- 
Mitchell Jon Stanton-Cook
PhD Student
Biomolecular modelling group
University of Queensland

m.stantoncook at uq.edu.au