[gmx-users] Atom positions change between topology generation and NVT equilibration

[Max Winokan]

Dear GMX-Users,

 

I am running some simulations of unsolvated DNA base pair dimers, with a topology generated from a structure previously optimised with quantum mechanical models. Because these structures have been highly-optimised it is important that Gromacs begins the MD with these atom positions.

 

After generating the topology with pdb2gmx I get the structure file (gc.gro) with the configuration seen below (after_pdb2gmx.jpg):

 

 

This configuration matches the optimised input structure. However, when I examine the trajectory (.trr) produced by running my NVT simulation (the MDP file for which is attached), the first timestep is very different from the input gc.gro structure, as seen below (first_nvt_frame.jpg):

 

 

The discrepancy is largest with the hydrogen atoms (which are not constrained of course), but all the atoms have been slightly shifted from the input configuration. Please could someone fill me in on the source of these position changes, and how I can modify my methods to avoid these.

 

Thank you and best regards,

 

Max Winokan

 

PhD Candidate in Theoretical Physics

Leverhulme Quantum Biology Doctoral Training Centre University of Surrey

GU2 7XH Guildford

United Kingdom

Office: 03AZ04

Email: m.winokan at surrey.ac.uk

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20200429/344164ce/attachment-0001.html>