[gmx-users] Question about Berendsen thermostat and Nose-Hoover temp coupling

[Justin A. Lemkul]

Chris,

I too rely heavily on the Berendsen methods for T- and P-coupling, and have 
always found them satisfactory.  I have not seen much beyond the casual 
references to the fact that N-H T-coupling and Parrinello-Rahman P-coupling are 
superior in terms of membrane simulations.  I have tried both, and have found 
both setups to be equally reasonable in their results.  It seems, overall, that 
Nose-Hoover schemes are often applied using NAMD and CHARMM software.  I don't 
know if this choice is motivated by the software, or the force fields used.

There are a few casual references to Berendsen being less than optimal in the 
literature and in archived presentation materials (which I can probably unearth 
if I can recall my Googling :).

Based on experience, I too would agree that Berendsen suffices, and only made 
mention of the N-H argument as a talking point, something that others have argued.

If I come across any of these references again, I will be sure to post them to 
the list to continue the discussion.

-Justin

chris.neale at utoronto.ca wrote:
> Hi Justin,
> 
> I wonder if you could expand upon the following statement, or perhaps 
> offer some links or references.
> 
> "Some argument can be made that N-H is more applicable to membrane 
> simulations."
> 
> I am interested because I use a Berendsen thermostat for membrane 
> simulations. To be entirely honest, I have no particular understanding 
> of the consequence of this choice beyond the oft-heard statement that 
> Berendsen T-coupling doesn't yield the correct ensemble. But it is very 
> intuitive and, as you say, it's usage is quite common and I have thus 
> far relied on these points to guide my choice of temperature-coupling 
> algorithm. Therefore I am hoping to take advantage of this opportunity 
> to learn a bit more about temperature coupling as you seem to understand 
> it quite well.
> 
> I am hoping that you can outline the logic underlying the 'correct 
> ensemble' statement and indicate how this is especially applicable to 
> membrane simulations.
> 
> I do realize that this is a large request on this type of mailing list. 
> Please feel free to decline.
> 
> Much obliged,
> Chris.
> 
> -- original message --
> 
> I have used the Berendsen method almost exclusively in my simulations, 
> and it is
> widely used in most of the literature I read.  While the argument can be 
> made
> that Nose-Hoover gives a result closer to the true ensemble than 
> Berendsen, I
> think both are sufficient for simple protein in water simulations.  Some
> argument can be made that N-H is more applicable to membrane simulations.
> 
> As for whether or not you need to slowly increase the temperature, it is 
> really
> up to you.  Usually such rigor is not necessary, as often position 
> restraints
> are applied to the protein during initial equilibration.  I have never
> experienced any adverse effects of simply setting the initial temperature,
> although that's not to say it's not possible :)
> 
> Just my $0.02.
> 
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use thewww 
> interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
> 

-- 
========================================

Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================