Thanks for your answer Thomas. Yes, there is a reason why I need to use
frozen and not just harmonically restrained atoms. It has to do with a kind
of new method for calculating free energies; the deltaF of part of the
system (part 1) was already calculated using that new method. Now the deltaF
for the rest of the system (part2) needs to be calculated using TI, but
keeping part1 completely frozen, or else there will be some kind of double
counting.
So I'll go on with this but will try to also do it somehow else, to be able
to compare and validate results.
Thanks again,
Ignacio
On Mon, Apr 13, 2009 at 12:46 PM, <steinbrt.rci.rutgers.edu> wrote:
> Hi,
>
> > Does anybody know why in the Amber code ibelly=1 (frozen atoms) and
> icfe=1
> > (thermodynamic integration) are incompatible?
> ...
> > I changed the code to allow this combination of parameters to run and, so
> > far, I haven't seen anything strange, apart from the fact that you have
> to
> > be careful and patient to get to a well equilibrated system.
>
> My personal guess would be that TI and ibelly can coexist as long as the
> belly region is exactly the same in both processes and doesn't contain
> changing parameters. However, the warning you disabled certainly means
> that no one ever tested this assumption and moreover that the TI code was
> written without the belly option in mind. The correctness of your results
> is therefore absolutely *not* ensured.
>
> Is there a reason you have to use ibelly instead of harmonic restraints?
> The later would not only be better tested but in my (and probably many
> other peoples) opinion a more realistic description of a restrained
> system.
>
> Kind Regards,
>
> Thomas
>
> Dr. Thomas Steinbrecher
> BioMaps Institute
> Rutgers University
> 610 Taylor Rd.
> Piscataway, NJ 08854
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Apr 15 2009 - 01:10:25 PDT
