[gmx-users] REMD - low temperature ranges

[Mark Abraham]

On 27/04/2012 10:59 PM, Tomek Wlodarski wrote:
> Hi,
>
> I have notice that quite often people in REMD simulation use replicas 
> in lower than 300K temp.
> Using for example temperature ranges from 250 to 450K
> I am wondering what is the purpose of those replicas.
> I have limited computational resources and I am wondering if for 
> studing 175 aa protein is better to start from 250K and up to 350 K or 
> choose from 300-400K (just an example)
>

Depends on the temperature(s) at which you wish to make observations. 
Sampling is normally enhanced at higher temperatures, so unless you want 
the ensemble at some low temperatures, it is normal to start at your 
lowest observation temperature.

Of course, you could always read the reasoning provided by the authors 
in their paper, or email them for clarification...

Mark