AMBER Archive (2008)

Subject: Re: AMBER: how to ger resp charge for a residue?

From: FyD (fyd_at_q4md-forcefieldtools.org)
Date: Fri Oct 31 2008 - 08:21:13 CDT

Hi Francesco,

> May I ask if the project of deriving RESP charges with NWChem is going
> on? We NWChem users hope it is going on. Could you take advantage from
> the NWChem built in method to derive ESP and RESP charges? I used it
> successfully some time ago, but then get discouraged by the complex
> procedure of putting that to use with Amber.

I remember you are interested in NWChem + RESP ;-) I am also
interested in extending R.E.D. to NWChem. However, as I already told
you this is not only interfacing NWChem + RESP, but showing that using
NWChem + RESP we get the same charge values than with Gaussian + RESP
or GAMESS-US + RESP or PCGAMESS/Firefly + RESP.

With Gaussian + RESP, or GAMESS-US + RESP, or PC-GAMESS/Firefly + RESP
the charge values are perfectly reproducible ONLY if you use R.E.D. &
the rigid-body algorithm implemented in R.E.D. Otherwise, the charge
values generated are not be reproducible. I would like to achive the
same results with NWChem and other QM packages, or demonstrate why the
results are different.

We will release data from q4md-forcefieldtools soon. I hope you will
find some features interesting for you, as a NWChem user.

> Could you also envisage that RESP charges for unusual ligands that do
> not covalently bind to macromolecules could be used for docking
> programs?

I do not see a difference in the strategy of deriving charge values
for a molecule covalently bound to another one or not covalently
bound. At the end you just get a Tripos mol2 file with connections or
without connection.

regards, Francois

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu