AMBER Archive (2003)

Subject: Re: PME

From: darden (darden_at_gamera.niehs.nih.gov)
Date: Wed Jun 11 2003 - 10:21:25 CDT


Dear Yuguang
I have an idea what may be the reason. I don't know for sure what is the
implementation of pme in gromacs. I sent them the spme code. This is
essentially what was in the original amber implementation. Since then I
have modified the B-spline weight term. This modification gives somewhat
more accurate energies but about the same accuracy for forces.

There is a flag in &ewald input section called opt_infl
It is set to one be default. Try setting opt_infl=0 and see if the ewald
energy now matches gromacs
hope this helps
Tom Darden
 On Tue, 10 Jun 2003, Yuguang Mu wrote:

> Dear David,
> I try to install AMBER force field to another MD program,e.g. GROMACS.
> When I compare two results,one from original AMBER, one from GROMACS,
> in the case of vacuum, cutoff, I got complete same results.
>
> But when I use PME, the results do differ. Mainly in the energy terms
> of electrostatics.
>
> I remember that in AMBER there are other switches related with PME,
> for example, VDWMETH. DO you think these kinds of switches can effect
> the simulation results greatly ?
>
> Do you think which system can be reliabley used to check such
> force field implimentations ?
>
> Thanks
>
>
> Dr. Yuguang Mu
> Institute for Physical and Theoretical Chemistry
> J.W. Goethe University Frankfurt am Main
> Marie Curie Str. 11
> 60439 Frankfurt/Main, Germany
> Tel: +49-(0)69-798-29711
>
> On Tue, 10 Jun 2003, Bimo Ario Tejo wrote:
>
> > Thanks Dr. Yuguang,
> >
> > Unfortunately, my output trajectory is coordinates file, not velocity. Is it another way to observe the helix/domain movement quantitatively without velocity information? FYI, the helix creates a flap over the protein which is open and closed during catalysis.
> >
> > Best wishes,
> >
> > Bimo
> > ITB - Stutgart
> >
> > ----- Original Message -----
> > From: Yuguang Mu <ygmu_at_theochem.uni-frankfurt.de>
> > Date: Tue, 10 Jun 2003 12:47:42 +0200 (MEST)
> > To: Bimo Ario Tejo <bimo7_at_linuxmail.org>
> > Subject: Re: helix movement
> >
> > > My rough idea is following :
> > > get coordinates and velocities of the helix/domain
> > > then calculate the exectly the overall translational and rotational
> > > velocities. then you may have a idea of such movement.
> > >
> > > Dr. Yuguang Mu
> > > Institute for Physical and Theoretical Chemistry
> > > J.W. Goethe University Frankfurt am Main
> > > Marie Curie Str. 11
> > > 60439 Frankfurt/Main, Germany
> > > Tel: +49-(0)69-798-29711
> > >
> > > On Tue, 10 Jun 2003, Bimo Ario Tejo wrote:
> > >
> > > > Dear Amber users,
> > > >
> > > > Does anyone have a tool to calculate the movement of helix and/or domain from Amber trajectory file? Superimposing some snapshots is useful, but it is not a quantitative way to describe such a movement.
> > > >
> > > > Any help is highly appreciated.
> > > >
> > > > Best regards,
> > > > Bimo
> > > > ITB - Stuttgart
> > > > --
> > > > ______________________________________________
> > > > http://www.linuxmail.org/
> > > > Now with e-mail forwarding for only US$5.95/yr
> > > >
> > > > Powered by Outblaze
> > > >
> > > >
> > >
> >
> > --
> > ______________________________________________
> > http://www.linuxmail.org/
> > Now with e-mail forwarding for only US$5.95/yr
> >
> > Powered by Outblaze
> >
>