[gmx-users] total simulation time for proteins in general

[Mark Abraham]

Priya Chandran wrote:
> Hello,
> I am a beginner in GROMACS and I am using GROMACS to analyse the 
> solution structure of protein in water after initial homology/abinitio 
> modelling. I have already run a search on the gmx_users list for my 
> queries. There are a few basic questions i would like to ask.

Welcome.

> 1. In the  md integrator, what is the total simulation time used in 
> simulation of peptides in general? I'm currently using 200ps, consisting 
> of nsteps=100000, and dt=0.002 and the rmsd graph has not reached a 
> plateau region still.

This is totally dependent on the system in question and what you expect 
the time-scale of relevant motions to be. I'd expect 200ps is at least 
an order of magnitude too short to have any basis for confidence that a 
solution structure is stable for a given forcefield.

> 2. Is minimisation in GROMACS only used as a precursor to the 
> md-simulation to remove bad contacts, or can we also use it effectively 
> to minimize a raw modelled ab-initio structure.?

Well, yes, you can use it for this - they're both the same objective, 
really.

Mark