So I couldn't wait and applied the patch from the bugreport that you have
linked. It did solve my current issues. Now I can see the update.2 is
available but is it OK to install it on top of leap.patch or should I start
fresh with unpatched code?
Thanks for all the input.
Regards,
Murat
On Wed, May 14, 2014 at 3:25 PM, Jason Swails <jason.swails.gmail.com>wrote:
> On Wed, 2014-05-14 at 02:02 +0300, MURAT OZTURK wrote:
> > Thank you very much for clarification. The bug explains the
> inconsistency I
> > was 'sensing' in the files. I have NATOMS 86347 and then 43150 lines, for
> > example. I will be recompiling in the morning. How would I make sure the
> > patch is upstream?
>
> The patch is upstream. See http://ambermd.org/bugfixesat.html (update.2
> for AmberTools 14).
>
> Amber/AmberTools has an automatic updater script that manages patches as
> well. If you just recompile Amber, the configure script will look
> online for available patches and ask you if you want to apply them if
> any are found. Simply say "yes".
>
> >
> > I can hack C/C++ but I am hopeless with FORTRAN, thanks for the format
> > information too.
> >
> > By the way, I am not at all interested in vacuum minimization, I just
> don't
> > see why it shouldn't work. It should run, at least until it explodes or
> > something, it is merely meant as a sanity check on my side.
>
> It doesn't work because it is not coded in pmemd.cuda. It may seem
> rather striking, but in reality there are no electrostatic methods that
> closely resemble vacuum conditions that Amber uses. The EEL functional
> form for PME electrostatics is _completely_ different (and imposes
> periodicity), as is the GB approach. None of these methods can really
> be implemented efficiently via "first calculate vacuum electrostatics,
> then add on a non-vacuum correction term" (at least I think), so
> including support for vacuum calculations would actually require an
> extra bit of coding.
>
> Given the numerous problems with simulating in vacuo, and the fact that
> vacuum calculations are supported in sander, no effort was made to
> implement it in pmemd.cuda, AFAIK.
>
>
> > As for placing ions, I find it preferable to have some concentration of
> > ions rather than the mere number of ions required to neutralize the
> system.
> > Given that, random placement is fine, so addionsrand use is preferable.
> > Thanks for that tip too.
>
> I typically do that as well. In that case, you can specify the number
> of ions you want to use. So for a 10- protein, for instance, I would do
> something like:
>
> addIonsRand system Na+ 40 Cl- 30
>
> to add 40 sodiums and 30 chlorides. (Note that "addIonsRand" works by
> replacing solvent, so you must always solvate first).
>
> 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 Wed May 14 2014 - 06:00:04 PDT
