[gmx-users] Re: mdrun on GROMACS 3.3.1

[Sheeba]

Hi, I have been working on the reverse transformation and I had the same
problem with mdrun generating the error:

Can not open file:
coarse.xtc

However this happens only when I use mdrun from the modified 3.3.1 version
without the '-coarse' option, which is similar to what Francesca vitalini
had used:

$MYGROMACSPATH/bin/mdrun -v  -s dynamin_dimer_PR1.tpr  -o
dynamin_dimer_PR1.trr  -cpo dynamin_dimer_PR1.cpt -c dynamin_dimer_PR1.gro

>From what I understand, the modified version is built to couple the CG
structure with the transformed AA structure using the parameters listed in
the input file. However after simulated annealing and releasing the
restraints when we try to run a regular MD run, there is no need to couple
with the CG structure and hence we ignore the '-coarse' option and mdrun
crashes. 

Ideally once the restraints are released the resulting .gro file can be run
with a newer version of Gromacs with the necessary updates to the itp files
where applicable. This I have done for my system and I am able to
successfully run a transformed AA system. 

However I also tried to find a work around to equilibrate the AA structure
with the mdrun in the modified 3.3.1 version and this can be achieved by
generating the tpr file for mdrun after removing the mapping section in the
all-atom itp files and then using mdrun with the '-coarse' option mentioning
the CG structure. This only gives an illusion that mdrun couples the CG
structure but since there is no mapping section in the itp files it does not
actually couple the CG structure and the resulting PE I get are similar to
the PE for a regular AA system. 

This is however my observation and I would appreciate if the developers of
the modified 3.3.1 version could check and confirm it. 

Sheeba





-----
Sheeba J. Irudayam
Postdoctoral Researcher
Department of Chemistry
University of North Carolina at Chapel Hill
isheeba[at]email.unc.edu
--
View this message in context: http://gromacs.5086.n6.nabble.com/mdrun-on-GROMACS-3-3-1-tp4458897p5000615.html
Sent from the GROMACS Users Forum mailing list archive at Nabble.com.