[gmx-users] Topolbuild

[Ray, Bruce D]

On Mar 21, 2018, at 10:15 AM, Alex <alexanderwien2k at gmail.com> wrote:

> Date: Wed, 21 Mar 2018 09:47:06 -0400
> From: Alex <alexanderwien2k at gmail.com>
> To: gmx-users at gromacs.org
> Subject: [gmx-users] Topolbuild
> Message-ID:
> 	<CAEDv6u7RpOKSq3WunOA7HweYChZODnEXv1LV2MHb=KFF936iEA at mail.gmail.com>
> Content-Type: text/plain; charset="UTF-8"
> 
> Dear all,
> 
> Using topolbuild the *.gro and *.top file for a GPS molecule (C_6H_14O_5Si)
> were generated. In the GPS molecule there is a --Si(OH)3 group in which the
> Si and H are indeed the atoms which produce problem in the topol file.
> The Si and that H are named like below in the [ atom ] section ;
> 
> [ atoms ]
> ;  nr    type   resnr   residu   atom   cgnr    charge      mass
>    1    opls_154     1       GPS       O    1   0.00000  15.99940
> ;   0.0000000
>    2         H??     1       GPS       H    1   0.00000   1.00800
> ;   0.0000000
>    3          SI     1       GPS      Si    1   0.00000  28.08000
> ;   0.0000000
>    4    opls_136     1       GPS       C    2   0.00000  12.01100
> ;   0.0000000
>    5    opls_140     1       GPS       H    2   0.00000   1.00800
> ;   0.0000000
> 
> And then the bond and angle sections wherever those two atoms are involved
> are empty like below;
> [ bonds ]
> ;   ai  aj   funct      b0          kb
>      24    23   1     0.10880      284512.       ;     H-     C
>      4     6   1     0.10900      284512.       ;     C-     H
>       4     3                                    ;     C-    Si
>      14    13                                    ;     H-     O
>      13     3                                    ;     O-    Si
>       3     1                                    ;    Si-     O
>       3    15                                    ;    Si-     O
> 
> Any suggestion to overcome the problem would be highly appreciated.
> 
> Best regards,
> Alex


You are using OPLSAA force field.  The versions of Gromacs' oplsaa.ff that I have seen
do include Si as indicated (see ffnonbonded.itp in oplsaa.ff).  However, bond, angle, and
dihedral parameters for Si are not present, nor are any atom types for O bound to Si
or H bound to O bound to Si.  You will have to use a text editor to manually edit those in
according to other information about what these atom types and parameters ought to be
because topolbuild lacks information about them and does not invent new parameters or
new atom types, but only tries to match bonding situations within submitted molecules to
descriptions of already defined atom types and then select whatever parameters might be
available for the situation.


-- 
Bruce D. Ray, Ph.D.
Associate Scientist
IUPUI
Physics Dept.
402 N. Blackford St., Rm. LD-061
Indianapolis, IN  46202