[gmx-users] parametrization of GABA for charmm27

[Justin Lemkul]

On 12/15/14 5:09 PM, jkrieger at mrc-lmb.cam.ac.uk wrote:
> Hi,
>
> I have parametrized gamma-aminobutyric acid (GABA; three-letter residue
> code ABU) for charmm27, which lacked it. This is essentially a hybrid of
> glutamate and lysine parameters. I thought this would be a reasonable
> starting point especially seeing as there is an overlap point at Glu CB
> and Lys CD, which share a charge of -0.18
>
> I would like to add my changes to the topologies archive
> (http://www.gromacs.org/Downloads/User_contributions/Molecule_topologies)
> for the benefit of other users but the "attach file" link is greyed out.
> Instead I copy them into this email.
>
> There are, however, some problems where pdb2gmx changes things (at least
> version 4.6.1 which I am using):
> 1. The charges of the carboxyl group are changed such that the topology
> doesn't end up with 0 charge. This needs to be manually corrected at
> present.
> 2. It also adds 3 extra hydrogens to the amino group (a total of 6 rather
> than 3), which also have to be removed manually from the topology and
> output structure. This requires additionally that the atoms be renumbered
> to continue consecutively.
> 3. As GABA lacks a carboxyl group on the amino carbon (CA) it has two
> hydrogens, which are correctly added. However the topology says that it is
> CT1 rather than CT2. This needs to manually corrected too, especially
> seeing as no N-CT1-H Urey-Bradley parameters exist.
>
> Here are the lines I added into aminoacids.rtp
>
> [ ABU ]
>   [ atoms ]
>          N       NH3     -0.30   0
>          H1      HC      0.33    1
>          H2      HC      0.33    2
>          H3      HC      0.33    3
>          CA      CT2     0.21    4
>          HA1     HA      0.05    5
>          HA2     HA      0.05    6
>          CB      CT2     -0.18   7
>          HB1     HA      0.09    8
>          HB2     HA      0.09    9
>          CG      CT2     -0.28   10
>          HG1     HA      0.09    11
>          HG2     HA      0.09    12
>          C       CC      0.62    13
>          OT1     OC      -0.76   14
>          OT2     OC      -0.76   15
>   [ bonds ]
>          CB      CA
>          CG      CB
>          C       CG
>          OT2     C
>          N       H1
>          N       H2
>          N       H3
>          N       CA
>          CA      HA1
>          CA      HA2
>          CB      HB1
>          CB      HB2
>          CG      HG1
>          CG      HG2
>          C       OT1
>   [ impropers ]
>          C       CG      OT2     OT1
>
> and here are the lines I added into aminoacids.hdb
>
> ABU     4
> 3       4       H       N       CA      CB
> 2       6       HA      CA      N       CB
> 2       6       HB      CB      CA      CG
> 2       6       HG      CG      CB      C
>
> Perhaps someone can see from these lines what is causing the strange
> problems above..
>

Most of your issues probably arise from the .tdb files; if this residue is being 
treated as an amino acid, it will have patches applied.  Without your input 
files and your actual commands, there's little to suggest.  Since your 
nomenclature overlaps with what is normally used for termini, this is probably 
your issue.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================