[gmx-users] extended bonds

[Justin Lemkul]

On 4/2/14, 2:31 PM, pratibha wrote:
> Hi all
>
> I have done simulations of my protein in water-cosolvent system (random
> initial addition) at 450K. At the end I have used:
> trjconv -f *.xtc -s *.tpr -center -pbc whole -o new.xtc
> But when I view my trajectory in vmd, I can see extended intramolecular
> bonds between cosolvent molecules. This problem usually come when pbc is not

Your explanation is not clear.  If something is intramolecular, it by definition 
is not between molecules, it is within one molecule.  Please clarify - are you 
seeing distorted bonded geometry or bonds that appear to be between molecules. 
If the latter, please see 
http://www.gromacs.org/About_Gromacs/Related_Software/Visualization_Software#Topology_bonds_vs_Rendered_bonds 
- bonds can't be created or broken during classical MD.  Most visualization 
software tries to guess bonded geometry based on distances; sometimes it guesses 
wrong.

> taken care off. But in my case I have used trjconv (as above) to account
> this. Is there something wrong in my simulations? I have never encountered
> such problem before at lower temperature of 300K.

If the problem truly is intramolecular, likely the topology is not fully stable 
at elevated temperature.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================