[gmx-users] RMSD, two simulations

[Xavier Periole]

On Thu, 26 Jun 2008 13:38:30 +0200
  Michal Kolinski <mkolin at iimcb.gov.pl> wrote:
> Dear Users,
The closest tool to your need is g_rms using the option -m, which
generates a matrix of rmsd between all pairs of conformations
present in the trajectory.

To compare the two trajectories you should concatenate them (trjcat)
into one unique trajectory.

To overcome the problem of different chain length (which has to be
resolved before concatenating!) you can use trajectories of subset
of proteins, namely using the common part between the two topologies.
trjconv can help you doing this with an index (different for each
model).
  
Retraining your analysis to trajectories of Calphas or backbone
atoms would speed up all the calculations.
  
> I've created two different models of the same 
> transmembrane protein.  These two models were 
> created using homology approach using two different 
> templates (both with similar sequence similarity 
> to the sequence of the target protein).  Next, I 
> did two separate 25ns simulations of those two 
> obtained protein models in water and lipid environment.  
> RMSD value calculated using g_rms reaches about 3.5 A 
> after 10 ns in two simulations and stays stable for 
> the rest of the MD time.  
> 
> What I want to do is to obtain a plot  of  RMSD 
> between those two models (model one vs model two) 
> for each corresponding frame of those two MD simulations
> (I just want to add that the length of protein chain is 
> different in both protein models).    Is there a script 
> available from gromacs tools which could be used for this?  
> Could you give me some sugestions how to treat this problem? 
> Thank you in advance.
> Michal 

-----------------------------------------------------
XAvier Periole - PhD

NMR & Molecular Dynamics Group
University of Groningen
The Netherlands
http://md.chem.rug.nl/~periole
-----------------------------------------------------