[gmx-users] Re: Nose-Hover chains for membrane protein simulation

[James Starlight]

Performing 5ns simulation after 2 ns equilibration in NPT ensemble (MTTK
barostat 5ps coupling ) I've observed non-physical behaviour of my system
with the constant drift of the protein molecule as the rigid body in the
y-z plane

Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Pressure                   -760.137         --    193.776    218.059  (bar)


>From manual I've noticed that MMTK doest not support *semiisotropic
scalling.  *Doest it means that with the Nose-hover chains and md-vv I
should use only weaker coupling during productions runs (I cant use
Parinello;s barostat with such options too)

Thanks for help

James



2013/5/31 James Starlight <jmsstarlight at gmail.com>

> Dear Gromacs users!
>
> I'd like to perform simulation of the membrane protein in lipid-water
> system using Nose-Hover with chains.
>
> From manual I've found that with such thermostat I should use (1) md-vv
> integrator (2) MTTK  instead of Parinello's batostat  and (3) shake instead
> of LINCS.
>
>
> How doest such options compatible with the simulation of membrane proteins
> in general ? On what other options should I pay attention during simulation
> of membrane protein with NH chains ?
>
>
>
> Thanks for help,
> James
>