Hi Murat and Jason
Thanks for your replies
That was nice suggestion Murat and I should try that as I am also not able
to find the reason for this great number of difference in water molecules.
and Also ...
>> However, you can always experiment with Amber and keep making the buffer
size a little larger until you get _almost_ the number of water molecules
you want, plus a small number more.
But the problem now is which one to believe Gromacs' or Amber's number of
water :( I ran initial job with AMBER and want to compare with Gromacs and
now found such a great difference with Gromacs. Changing the buffer size to
get same water molecules would be a problem in reporting in future. Say
with AMBER I used 12 Angs buffer and with Gromacs I used 8 Angs buffer to
keep waters same. That does not sound correct to report in publications :(
>> Since I don't know any specifics about gromacs, I don't know how to specifically
help
I had put same query in Gromacs archive and they also dont know what AMBER
does but I am posting here what Tsjerk Wassenaar from Gromacs replied.
Hope it helps to find the reason.
Gromacs query title:"" truncated octahedron box vector angles and number of
waters""
""
.... As for the difference
in water molecules, I don't know what AMBER does, but I can tell what
Gromacs does. You have the diameter of the system, 15.385 nm, to which <<<#
diameter is basically the longest length #>>>
twice the 1.2 nm is added, giving the vector length 17.785 nm. A cube of
that length would have a volume of 5625 nm^3, and a truncated octahedron is
about 78% the volume of a cube, so it gives 0.78*(17.785**3) = 4388 nm^3.
Well, there's some rounding..., the 4330.63 will be correct. Solvating that
with genbox should give a pretty much correct amount of water. If AMBER
gives you so much less, I'd guess there's something odd there.""
Thanks
On Sun, Jun 15, 2014 at 6:07 AM, Jason Swails <jason.swails.gmail.com>
wrote:
> On Fri, Jun 13, 2014 at 9:04 AM, newamber list <newamberlist.gmail.com>
> wrote:
>
> > Hi All
> >
> > I am solvating same system with truncated octahedron box and want to run
> > with both AMBER and GROMACs. In AMBER, am using 12 Ang buffer with
> > solvateoct and get around 76716 water molecules but with Gromacs with
> same
> > buffer I am getting around 132902 waters (almost twice). These are
> > following commands am using:
> >
> >
> > with AMBER:
> > solvateoct sol TIP3PBOX 12
> >
> > with GROMACS:
> > editconf -f input.gro -o output.gro -c -d 1.2 -bt octahedron
> >
> >
> > I am not sure whats wrong, please let me know if someone got similar
> > issues?
> >
>
> I'm not familiar with Gromacs at all. However, you can always experiment
> with Amber and keep making the buffer size a little larger until you get
> _almost_ the number of water molecules you want, plus a small number more.
>
> Then you can use ParmEd or cpptraj to create a new topology file by
> stripping away the "extra" water molecules.
>
> In order for GROMACS and Amber to be getting that big of a difference in
> the total number of solvent molecules, they must be doing different things.
> Since I don't know any specifics about gromacs, I don't know how to
> specifically help.
>
> HTH,
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
> _______________________________________________
> 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 Sun Jun 15 2014 - 07:30:03 PDT
