VMD-L Mailing List

From: Paweł Kędzierski (pawel.kedzierski_at_pwr.wroc.pl)
Date: Sun Jun 12 2011 - 08:59:40 CDT

On 12.06.2011 12:57, sarah k wrote:
> Dear Axel and Hamzeh,
> Thanks for your replies. We have put our protein in nitrogen drift
> gas, experimentally and now we need to simulate it.
>
> how high of a pressure of nitrogen do you expect to model?
>
> *P=* 1 atm
>
> how is the nitrogen supposed to impact the protein, if at all?
>
> The protein is ionizined by high voltage in the drift gas.
>
> technically, it is of course possible, but before you do this,
> please do a back of the envelope calculation:
> - assume the nitrogen is an ideal gas
> - the desired pressure and volume of your gas box
> - calculate how many nitrogen atoms would be in that box
>
> *N=* 10^28 molecules, T= 438 K, V= 0.085 L
Given P, T and V this is about 1.5*10^21 molecules.
Even with that you have to scale down your expectations unless you have
something like 10^15 GB of RAM and about as many CPUs...

> The final result may differ signiciently if I suppose the protein in
> water or vacuum.
The forcefield approach is not well suited for such simulations. You
could model protein in any given ionization state but you can't change
it (well, easily) during the simulation. And unless you include ionized
nitrogen explicitly, the interaction with the gas will not depend in any
way on the charge distribution on the protein, since polarization is not
modelled. Apart from that, the constant bombardment of the protein with
gas molecules could be much more effectively modelled implicitly (that
is, in vacuum) using Langevin dynamics.
Regards,
Pawel K.

> Regards,
> F. Keshavarz