[gmx-users] custom topology questions

Justin Lemkul jalemkul at vt.edu
Wed May 2 03:51:12 CEST 2018



On 5/1/18 5:40 PM, Eric Smoll wrote:
> Hello Gromacs Users,
>
> I have a few custom topology questions. Input on any of these questions
> would be greatly appreciated.
>
> 1.) What gromacs tools use the information in atomtypes.atp?

pdb2gmx only

> 2.) Table 5.4 claims the atomtypes directive consists of lines with the
> format:
> atom-type, mass, charge, particle-type, V-parameter, W-parameter
> in the opls-aa directory, ffnonbonded.itp has:
> atom-type, bond-type, atomic-number, mass, charge, particle-type,
> V-parameter, W-parameter
> Is it OK to use either format safely? Gromacs isn't hardcoded to use a
> special format for the opls-aa directory, correct?

No, grompp can read a variety of different formats that are different 
between force fields.

> 3.) consider "configuration A":
> gen-pairs is set to yes, fudge-LJ to 0.5, and fudgeQQ to 0.5, and a "pairs"
> directive is provided with two atom-numbers per line. I expect this
> configuration will calculate scaled 1-4 interactions (LJ and coulomb) and
> apply them to all the atom-number-pairs listed in the "pairs" directive.
> Now consider "configuration B":
> gen-pairs is set to yes, fudge-LJ to 0.5, and fudgeQQ to 0.5, and a "pairs"
> directive is provided with two atom-numbers, a "1" for the function type,
> an X-parameter, and a Y-parameter for every line.
> Am I correct in assuming the gen-pairs and fudgeLJ do nothing here? Am I
> correct in assuming that fudgeQQ is still applied to 1-4 coulomb
> interactions? When parameters are provided in the "pairs" directive, they
> only apply to LJ-type interactions, correct (i.e., the only way to tune 1-4
> coulomb interactions is with the fudgeQQ factor)?

For pairs of function type 1, yes, but with function type 2, you can 
control 1-4 electrostatics.

For both of these cases, you can compare the results of equivalent 
configurations and different topologies with mdrun -rerun.

> 4.) In the "dihedrals" directive, can the user specify a type-1 dihedral
> ("proper dihedral") repeatedly to form a superposition potential or do
> later parameters override previous parameters. For example,
> [ dihedrals ]
> ;i   j   k   l   function-type   phi_s, k_phi, multiplicity
> 1 2 3 4 1 0 1 1; (1 + cos(phi))
> 1 2 3 4 1 0 1 1; (1 + cos(phi))
> 1 2 3 4 1 0 1 1; (1 + cos(phi))
> to yield a potential equivalent to 3*(1 + cos(phi)) or (1 + cos(phi))?

Repeated parameters in this way override previous. You can only use a 
type 9 dihedral to add multiple terms, but those must have different 
multiplicities.

> Thank you!
>
> Best,
> Eric

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

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.thelemkullab.com

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



More information about the gromacs.org_gmx-users mailing list