[gmx-users] ffbonded.itp file

[Justin A. Lemkul]

anna.marabotti at isa.cnr.it wrote:
> Dear gmx-users,
> I’m trying to modify the ffbonded.itp file of Gromos43a1 ff in Gromacs
> 4.5.3 version, in order to add parameters for my modified Cys residue.
> However, I have some difficulties in understanding what are the
> parameters already included in this file. In particular, ffbonded.itp
> file appears as:
> 
> ; Table 2.5.2.1
> ;       GROMOS bond-stretching parameters
> ;
> ;
> ;	Bond type code
> ;	Force constant
> ;	Ideal bond length
> ;	Examples of usage in terms of non-bonded atom types
> ;
> ;
> ;	ICB(H)[N]    CB[N] B0[N]
> ;
> #define gb_1        0.1000  1.5700e+07
> ; H  -  OA      750
> ;
> #define gb_2        0.1000  1.8700e+07
> ; H  -  N (all) 895
> ;
> #define gb_3        0.1090  1.2300e+07
> ; HC  -  C      700
> ....................................
> 
> but while I understand the meaning of the first two numbers, I don’t
> understand what the number near the description of the two atom type is
> (for example, in the definition of gb_3, what is “700” after HC – C? I

These are index values, which I believe were ported from the Gromos software 
eons ago.  They have no meaning in the context of Gromacs (and you'll note that 
it's a comment, anyway).  The necessary parameters are all contained within the 
#define statements, which are then mapped back to the topology when invoked.

> looked at the Manual of Gromacs version 4.5.3, section 5, paragraph
> 5.3.3. Bonds (page 111) but the “example excerpt” for ffbonded.itp file
> is very different from the one I see in my ffG43a1 directory! I don’t

The Gromos force fields are somewhat unique.  The parameters are assigned based 
on #define statements.  All the other force fields follow the format shown in 
the manual, so the documentation reflects the most general case (which, as 
implied by the #defines, is still effectively maintained here).

> see in the manual or elsewhere a detailed explanation for the meaning of
> each component of this file. Could you please help me?
> Moreover, I see some strange definitions, at least in my opinion. I’m
> trying to mutuate parameters from DMSO: (CH3)2S=O. Under the section
> Bond type, I see these definitions:
> 
> #define gb_44     0.280412  2.3900e+06
> ; ODmso -   CDmso        900
> ;
> #define gb_45     0.292993  2.1900e+06
> ; CDmso -   CDmso        900
> 
> I really don’t understand the sense of the definition of these bonds
> types, provided that they are absent in DMSO since the two methyl groups
> are linked only to sulphur, not to each others (no C-C bond) and that
> oxygen is linked only to sulphur (no O-C bond). Could you please check
> for these strange interactions, or am I mistakenly interpreting them?

These are not bonds in the sense of true chemical covalent bonds, but rather 
they are used to fix the geometry of DMSO.  I am not aware of any publications 
on the 43A1 DMSO parameters (although 53A6 DMSO does have a publication, and 
since the same bonds are present you might investigate that particular paper...I 
think it was in J Phys Chem B), which makes it hard to confirm the purpose of 
these "bonds."  Presumably, since all Gromos solvent models are rigid, these are 
present to force some geometry, as is the case with, e.g. CHCl3.

-Justin

> Thank you very much and best regards
> Anna Marabotti

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

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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