Re: [AMBER] Incorrect ligand parameterization using GAFF

From: Hannes Loeffler <Hannes.Loeffler.stfc.ac.uk>
Date: Fri, 14 Apr 2017 09:23:26 +0100

On Thu, 13 Apr 2017 17:16:07 +0000
Emilia Pecora-de-barros <empecora.ucsd.edu> wrote:

> region of the frcmod file describing the dihedrals of interest:
> DIHE
> cc-ce-cd-cd 1 1.000 180.000 2.000 same as X
> -ce-ce-X nd-cc-ce-cd 1 1.000 180.000 2.000
> same as X -ce-ce-X nd-cc-ce-ha 1 1.000 180.000
> 2.000 same as X -ce-ce-X c -cc-ce-cd 1 1.000
> 180.000 2.000 same as X -ce-ce-X c -cc-ce-ha 1
> 1.000 180.000 2.000 same as X -ce-ce-X
> cd-cd-ce-ha 1 1.000 180.000 2.000 same as X
> -ce-ce-X

This looks like the output from parmchk. Parmchk2 gives you larger
barrier heights but also a large penalty score.

DIHE
cd-cd-ce-cc 4 26.600 180.000 2.000 same as X
-ce-cf-X , penalty score=136.0 nd-cc-ce-cd 4 4.000
180.000 2.000 same as X -ce-ce-X , penalty score=136.0
nd-cc-ce-ha 4 4.000 180.000 2.000 same as X
-ce-ce-X , penalty score=136.0 c -cc-ce-cd 4 4.000
180.000 2.000 same as X -ce-ce-X , penalty score=136.0 c
-cc-ce-ha 4 4.000 180.000 2.000 same as X
-ce-ce-X , penalty score=136.0 cd-cd-ce-ha 4 16.000
180.000 2.000 same as X -cc-cd-X , penalty score=136.0

Also, antechamber reports:

Info: Bond types are assigned for valence state 1 with penalty of 1
Total number of electrons: 130; net charge: -1

So some problem with assigning bonds and has your molecule really a net
charge of -1?


Cheers,
Hannes.

_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 14 2017 - 01:30:02 PDT
Custom Search