Actually, further digging revealed that the issue with the charge appears
initially in sqm.in and sqm.out. In fact, the total mulliken charge
reported in sqm.out is not zero, rather is something around 0.003 or 0.002
depending on molecule. Even the Sustiva sample molecule in the Amber
tutorial has a non-integer charge. So, then this non-integer charge
persists later in the mol2 file, Gromacs topology and Lammps data file.
Regards,
Salman Zarrini
---------- Forwarded message ---------
From: Salman Zarrini <salman.zarrini.gmail.com>
Date: Fri, Nov 13, 2020 at 2:44 PM
Subject: Antechamber: Charge and dihedral issues
To: <amber.ambermd.org>
Dear all,
Using Antechamber (-V 20) I am trying to generate GAFF parameters for a
neutral molecule to convert them later to Gromacs topology (*.top and
*.gro) files using tleap and parmed.
Below are the steps I take on an initial molecule's pdb file:
1. antechamber -at gaff -fi pdb -fo mol2 -i case.pdb -o case.mol2 -c bcc
-rn CASE -dr no
2. parmchk2 -i case.mol2 -o case.frcmod -f mol2 -a Y
3. tleap -f tleap.bat >>> to get the case.top case.crd
4. amb2gro_top_gro.py
I noticed the two below issues with charge and dihedrals on which your
comments are highly appreciated:
*Charge*: Some decimal points of charge (0.002998) in total appear in the
molecule making the structure charged and of course the decimal charge
would scale up by adding more molecules in the system.
The charge issue has nothing to do with converting to Gromacs format as the
charge issue persists even if I convert them to LAMMPS input files.
*Dihedral*: The dihedral are defined twice for some of the atoms as
following:
;ai aj ak al funct
1 2 7 9 1 0.00000 4.91620 2
1 2 7 9 1 0.00000 0.60250 3
or
3 1 2 7 1 180.00008 0.41840 2
3 1 2 7 1 0.00000 1.60247 3
or
33 34 40 37 1 180.00008 0.41840 2
33 34 40 37 1 0.00000 1.60247 3
Nevertheless, I admit the dihedrals have different multiplicity which might
get them to be counted as different dihedrals.
I tested several molecules and the issues persist, so, I would be so
thankful if you could let me know where I am doing wrong or some potential
factors that I should consider.
Best regards,
Salman Zarrini
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Nov 15 2020 - 11:00:02 PST
