[gmx-users] What is the purpose of the [ pairs ] section?

[Javier Cerezo]

Hi

The pair section specifies the atoms between which a non-bonded 
interaction will be build up, even if the "normal" non-bonded is 
excluded, what is a mechanism to build 1-4 interactions. So, if you want 
a 1-4 interaction to be taken into account, you need to specify it in 
the topology, in the same way you need to include all bonds, angles...

VdW pair interactions should be listed in ffnonbonded.itp, under [ 
pairtypes ] directive, where grompp can find them. However, if such 
interactions are not there, it is still possible to automatically 
generate them in the same way as "normal" VdW are generated, and 
additionally scaling by fudgeLJ. For such pair generation, gen-pairs 
should be set to yes in [ defaults ], otherwise grompp will give an 
error if it does not find them in ffnonbonded.itp. Note that if pairs 
are deffined in ffnonbonded.itp and you set gen-pairs to yes, they will 
be taken from ffnonbonded.itp.

As for coulomb pair interactions, they will be generated only through 
gen-pairs=yes, as (there is not any directive for them in 
ffnonbonded.itp), simply scaling the "normal" coulomb by fudgeQQ.

So, in the case of OPLS, the resulting 1-4 interactions, providing the [ 
defaults ] that are in forcefield.itp and the fact that there is not [ 
pairtypes ] for this force field, will be those prescribed for the force 
field, but only if you include the appropriate line under [ pairs ] for 
all possible combinations.

Javier

El 08/12/12 23:12, Andrew DeYoung escribió:
> Hi,
>
> I am sure that this is a rather basic question, but I am wondering if you
> can please help me to understand something described in the manual.
>
> What is the purpose of the [ pairs ] section of an .itp file?  What
> information does it contain?
>
> Section 5.3.4 (page 112) of the version 4.5.4 manual says:
>
> "Extra Lennard-Jones and electrostatic interactions between pairs of atoms
> in a molecule can be added in the [ pairs ] section of a molecule
> definition. The parameters for these interactions can be set independently
> from the non-bonded interaction parameters. In the GROMOS force fields,
> pairs are only used to modify the 1-4 interactions (interactions of atoms
> separated by three bonds). In these force fields the 1-4 interactions are
> excluded from the non-bonded interactions (see sec. 5.4)."
>
> However, I am using the OPLS force field, not GROMOS.  OPLS prescribes that
> 1-4 nonbonded interactions are to be taken at half strength (i.e., fudgeQQ =
> 0.5 and fudgeLJ = 0.5).  Does this mean that I should specify all possible
> 1-4 nonbonded interactions in the [ pairs ] section?  If I have nrexcl = 3
> specified in [ moleculetype ], then can I assume that 1-5, 1-6, 1-7, etc,
> interactions will be taken into account (at full strength), even if such
> pairs are not specified in [ pairs ]?  I have set gen-pairs = yes, fudgeLJ =
> 0.5, and fudgeQQ = 0.5.
>
> Thank you for your time!
>
> Andrew DeYoung
> Carnegie Mellon University
>

-- 
Javier CEREZO BASTIDA
Ph.D. Student
Physical Chemistry
Universidad de Murcia
30100, Murcia (SPAIN)
T: (0034)868887434