On Fri, Aug 30, 2013, Hamed S. Hayatshahi wrote:
>
> I am trying to do a simple TI (disappearing of a cation in a water
> box) with Amber12 sander using softcore potential. I have a couple of
> confusions:
>
> 1- The A9 tutorial sets up different crgmask values for V0 and V1
> processes, thus making different mdin files and also different starting
> coordinates, leading to a group file like this (from the liked shell
> script):
>
> -O -i mdin_min_v0_l${X} -o ${pre0}_min_v0_l${X}.out -p ${pre0}.prm -c ${pre0}.rst -r ${pre0}_min_v0_l${X}.rst
> -O -i mdin_min_v1_l${X} -o ${pre1}_min_v1_l${X}.out -p ${pre1}.prm -c ${pre1}.rst -r ${pre1}_min_v1_l${X}.rst
>
> But under the related topic, page 114 of the manual clearly says: "The
> input (mdin) and starting coordinate files must be the same for the
> two groups".
I agree that the manual is a little confusing, but a careful reading
of section 4.1 will help. The second paragraph there starts by saying
that there are two ways of proceeding: the first requires the prmtop
files to have the same number of atoms, etc. Later, this paragraph says
that "Alternatively," you can use the softcore algorithm **described
below**. The text that immediately follows (which you quote) then refers
to the first option. All further information about the softcore option is
"below" (in section 4.1.2).
The tutorial is using the softcore model described in Section 4.1.2, and you
should refer the the description there, which specifically talks about how
the two mdin files can differ (see the description of "scmask").
[Thanks for your report: we will make the manual clearer about the
distinction between the two alternatives in the next release. This will
be doubly important since there will then be yet more ways to do TI.]
> Also when I use different mdin and/or starting coordinate
> files, sander.MPI crashes with the following last lines in the output:
>
> | MPI Timing options:
> | profile_mpi = 0
We would need more information to help with this error. I don't think it
has anything directly to do with the fact that the mdin or coord files are
the same or different.
...regards...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 30 2013 - 13:00:02 PDT
