David -
I have been attempting to help Guillermina with this one. The molecule
framing information is from the nsp() array in the prmtop; the residue
information is from the ipres() array in the prmtop (these have different
names in pmemd, but that is what they are). To my knowledge, this stuff
affects 1) ekcmt calculation, 2) molecular virial calculation, and 3)
wrapping. Items 1 and 2 effect the pressure calculation, and from my
experience, slight differences actually are noticeable (based on what
happens when you define the molecular virial slightly differently, as in
sander 6 vs. sander 7). I agree it probably does not matter that much
because the solvent effect is much larger. However, I would think that
wrapping could get a little sticky. There is actually significant
computational overhead associated with getting all this stuff right. This
stuff is important in pmemd because pmemd does atom division on molecule
boundaries for constant pressure simulations. This makes it possible to not
exchange velocity information between processors, except when the master
needs to print it. I would think we would want to be getting the molecule
and residue framing right. I asked Guillermina to carefully check how the
residues define connect0,1,2 atoms and how they are used before filing this
as a bug. I am going to add code to pmemd that confirms that the framing
makes sense (ie., molecules don't contain pieces of residues).
Regards - Bob
----- Original Message -----
From: "David A. Case" <case.scripps.edu>
To: <amber.scripps.edu>
Sent: Wednesday, September 24, 2003 2:45 PM
Subject: Re: AMBER: residues and molecules
> On Tue, Sep 23, 2003, GUILLERMINA L ESTIU wrote:
> >
> >
> > The problem is that (I don't know how) tleap properly build the
residues, but
> > messes up the definition of the molecules : a molecule ends in the
middle of a
> > residue.
>
> What variables are you looking at to decide where the molecules begin and
end?
> I think sander only cares about this for determining the pressure, and
that
> the differences between different "molecule" definitions probably have
> a negligable effect, as long as the waters are correctly identified (they
> dominate by far the pressure calcualtion).
>
> I can't tell from your e-mail what could have gone wrong -- you mention
> something about a "unit docked", but I did not understand that.
Generally,
> the input pdb file to LEaP needs to have a TER card at the end of every
> molecule...it's not clear to me whether or not that was the case for your
> example.
>
> Of course, there could be glitches in the way LEaP figures out molecules,
but
> one would need more information to delve into this.
>
> ..dac
>
> --
>
> ==================================================================
> David A. Case | e-mail: case.scripps.edu
> Dept. of Molecular Biology, TPC15 | fax: +1-858-784-8896
> The Scripps Research Institute | phone: +1-858-784-9768
> 10550 N. Torrey Pines Rd. | home page:
> La Jolla CA 92037 USA | http://www.scripps.edu/case
> ==================================================================
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Sep 24 2003 - 20:53:01 PDT
