On Thu, Sep 15, 2011, Yongmei Wang (ywang) wrote:
>
> The AMBERtool manual 1.5, describes a command,
>
> set default Dielectric constant (or distance)
>
> I am pretty certain this command does not work. For I have tested with a
> tleap script file, with that argument set to constant or distance.. it
> produced exactly the same Parm files in both cases, and the same parm
> file when I delete that line from the script file. So.. could anyone
> confirm with me that this command is not valid? (I am using AMBER10)
This command (which is so little used we should probably retire it) determines
how addIons places ions into an electrostatic field. As you note, it doesn't
have anything to do with how subsequent programs handle electrostatics, and
hence has no effect on the prmtop file.
The manual does say that this pertains to how electrostatic calcuations *in
LEaP* are handled, but I agree that we could add comments to make this
clearer.
> Also under the next section when you explain eedmeth=5 option under
> Ewald summation. I would suggest you change the wording there. The
> distance-dependent dielectric you are using is the following function
>
> epsilon(r)=epsilon_0 * r; (epsilon(r) stands for distance dependent
> dielectric constant,epsilon_0 is whatever the dielectric constant
> user specifies). As a result, the electrostatic interaction has
> 1/r^2 dependence. This function is used for direct part of Columb
> Interaction. But I am not clear about the long-range part of
> Coloumb interaction (the part handled by PME). Does PME considers
> distance-dependent dieletric effect for the long-range electrostatic
> interaction?
As the manual states, "the last two options [eedmeth=4 and 5] are intended for
non-periodic calculations where no reciprocal term is computed." Your
suggested wording for the direct sum has some merit.
...thx...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Sep 15 2011 - 12:30:02 PDT
