[gmx-users] Different initial conditions for the membrane protein simulation

[James Starlight]

Mark,

1) Commonly I start my simulation with the V-rescale with the posres
applied on protein atoms ( this is equilibration phase) and after
stabilization of the temperature I switch to the N-H.

By the way I've found that the usage of chains N-H with the md-vv
integrator could significantly increase ergodicity of trajectory
without producing of the artifacts due to the wrong energy
distribution. I'm not sure about correct way of definition of the
number of chains for that setup and made separate topic about this
issue. By the way what addition problems could be arrisen with the
usage of that thermostat as well as md-vv integrator during simulation
of the membrane protein in NPT ensemble ?


2) Is there any way to create some energy diagrams for conformers (
snapshots taken in defined intervals) distribution along the
trajectory which could be indirectly compared with the projection of
that conformers onto some principal components?

James

2012/9/10, Mark Abraham <Mark.Abraham at anu.edu.au>:
> On 11/09/2012 2:36 AM, James Starlight wrote:
>> Dear Gromacs Users!
>>
>>
>> I'm simulating intrinsic dynamics of some membrane receptors which can
>> have functional relevance. Therefore I'm looking for the most suitable
>> initial conditions of my simulation which could provide maximal degree
>> of such dynamics (arisen from the cooperative thermal fluctuations for
>> instance). As I know the proper chosen thermostat can significantly
>> influence on the overall dynamics of the protein. In particular I've
>> found that Noose-Hover thermostat can promote it significantly. But
>> there are some theoretical studies which suggest that Nose-Hover
>> thermostat can produce alot of artifacts due to the incorrect energy
>> distribution on the micro-states. So the dynamics produced onto such
>> trajectories can be unphysical in general. That authors also suggest
>> use collision thermostat ( in GROMACS usage of SD integrator
>> correspond to that conditions without any t_coupl ).
>
> Or the V-rescale thermostat, which claims correct thermodynamics and
> kinetics, and is much easier to use than N-H.
>
>> So the main question- IS there any way to observe impact of each
>> thermostats and other initial conditions on the overall dynamic
>> behavior of the protein. For example does it possible to make some
>> energy landscape of the trajectories calculated from different
>> conditions to compare thermal energy distribution on the micro-states?
>> Finally is there any way to compare such energy landscapes with the
>> conformational maps ( e.g projections onto some lowest Principal
>> components) to direct monitor the energy contribution of each
>> conformation ?
>
> Not sure. The difficulty with comparing multiple setups is that in
> principle you need a converged simulation for each and some way to
> assess which is the best model of reality, and usually at least one of
> those is infeasible.
>
> Mark
> --
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>