AMBER Archive (2007)

Subject: Re: AMBER: RED-vIII fails

From: FyD (fyd_at_q4md-forcefieldtools.org)
Date: Tue Oct 30 2007 - 15:38:03 CST

Quoting Jesper Soerensen <lists_at_jsx.dk>:

> Files are attached, I figured I'd include all the data, as the file size
> wasn't too different.

- 1st in the tutorial
http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php, it is written
in the introduction:
"RESP was recompiled using qtol = 0.1d-5, maxq = 5000, maxlgr = 500
and maxmol = 200"
You are using a two molecules RESP fit; with for the 1st molecule (2
conformations x 4 orientations) and for the 2nd molecule (1
conformation x 4 orientations). Thus, your charge derivation start to
complex; I would recompile RESP with the keywords provided to avoid
additional error messages...

- Then, in http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php#11,
the INTER-MCC are defined between a molecule 1 & a molecule 2 (only).
The molecule 1 being the donor of chemical group and the molecule 2
the acceptor of chemical group. In your example, you inverted the
molecule 1 and 2 and used the molecule 2 as a donor of chemical group.
Try to switch the order of the molecule, and see how it works: It
should be OK.
I want to underline that the way INTER-MCC have been set up in
R.E.D.-III is quite rigid. In particular, R.E.D.-III only allows to
build molecular fragments as defined in Cieplak et al. J. Comput.
Chem., 1995, 16, 1357-1377. We changed that in R.E.D. IV and the
building of more complex force field topology databases is now
possible (and the way to do it is really less rigid).

- Finally, once you will have generated the Tripos mol2 file for your
terminal fragment you might considering reading
http://q4md-forcefieldtools.org/Tutorial/leap.php. We just put new
versions of the LEaP files to be downloaded.

Please, tell me if this solves your problem...

regards, Francois

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu