[gmx-users] How to use -tablep mdrun option?

[Mark Abraham]

On 27/04/2012 11:07 PM, Marcelo Lopez wrote:
> Hi all,
>
> I'm trying to run a coarse grained system with tabulated non-bonding
> potentials for 3 kind of particles. I need to set all the 1-4
> interactions to zero

Why not use nrexcl = 3 in your [moleculetype] like the example in manual 
5.7.1? (and probably gen-pairs = no for your force field)

>   but I can't find in the manual or elsewhere hot
> to construct the tables for the -tablep option of mdrun.

Searching the manual text for "tablep" finds me information in section 
7.3, which strongly suggests the information in 6.7.2 is applicable.

>
> I've readed a thread in the mailing list but still the information I
> get is uncomplete... (the manual says absolutely NOTHING!)
>
> 1) In these tables, do I specify the index number of every 1-4 pair or
> only the labes of the groups involved? Followed by the C6 and C12
> parameters.

Only a functional form is specified in the table. Topology (which 1-4 
pairs exist) is entered in the .top file. Parameters (charge, C6 and 
C12) likewise. Go and look at 6.7.2 and the example tables in your 
GROMACS distribution.

> For example, for the atom index 1 and 5 (energy groups A and B), one
> tablep*.xvg file must contain this
>
> 1   5   0.0   0.0
>
> or this (valid for all the AB 1-4 pairs?)?
>
> A    B    0.0    0.0
>
> I ask... And still something is missing?
> The system has no charges, do I say this too in every  tablep*.xvg file? how?

Charges are specified in the [atoms] section of your [moleculetype]. If 
they're zero, then nothing will ever be computed from them, so you can 
probably use anything you like with the the f() part of the tabulated 
potentials.

Mark