[gmx-users] High temperature simulations

[Erik Lindahl]

Hi,

No, all normal force fields are parameterised for liquid (that's the l 
in OPLS) simulations at room temperature. I really don't know of any 
work that has been done to systematically assess parameterisations at 
higher temperature; it is difficult enough at normal temperatures :-)

SPC/TIP3P are further very simple water models optimized for speed. The 
freezing temperature is actually almost perfect, only a couple of 
degrees off, but I can well imagine there are huge problems when you 
try to vaporize it. Jay Ponder is doing work on very advanced water 
models (including multipole interactions), but those are about two 
orders of magnitude slower to simulate. There are a couple of slightly 
more advanced models (compared to SPC/TIP3P) in the topology directory 
that you can play with too.

Cheers,

Erik



On Tuesday, Mar 4, 2003, at 03:20 US/Pacific, Oliver Beckstein wrote:

> Hi,
>
> may I add my question to the one Dallas asked:
>
> Are these force fields (ffgmx, GROMOS, OPLSA,...) actually parametrised
> for high temperatures (or more precisely: does one expect the
> parametrisation to be T-independent (and if yes, over which range)), 
> and
> how much can you trust the simulated behaviour of proteins or water at
> high T?
>
> (I know at least of two areas where this question may be important: 
> One is
> obviously the study of proteins at high temperatures, the other is use
> of high T (>500K) simulations in replica exchange simulations to allow 
> for
> better sampling of configuration space)
>
> Oliver
>
> On Tue, 4 Mar 2003, Dallas Warren wrote:
>
>> This is just a general question about high temperature simulations.  I
>> don't do them, and don't see that I will need to, but are simply 
>> curious ....
>>
>> Supposedly at 100oC, 373K, and 1 atm, water should boil.  Is it a
>> deficiency in the force-field for H2O that it doesn't (i.e. it isn't a
>> complete representation that exhibits this behaviour), or is 
>> something done
>> to make it stay as a "liquid"?
>>
>
> -- 
> Oliver Beckstein * oliver at bioch.ox.ac.uk
>  http://indigo1.biop.ox.ac.uk/oliver/
>
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.