[gmx-users] Adapting metallic ions from AMBER to GROMACS

[Pedro Lacerda]

Hi Gromacs users,

We are planning a simulation with a metallic (fe+2) attached on the
protein. The parameters found in [1] cannot be directly inserted on
ffnonbonded.itp because equation 1 of [1] and equation 4.5 of [2] are
slightly different.

Theoreticaly these Lennard-Jones parameters could be used in any PME based
force field, but it was derived and tested using the AMBER software and
potential, so it is our current force field of choice. In
amber99sb-ildn/ffnonbonded.itp the long range interactions are described in
terms of epsilon (kJ/mol) and sigma (Rmin; nm).

Eq 1:

    Lj = 4e ((R/r)^12 - (R/r)^6)

Eq 4.5:

    Lj = e ((R/r)^12 - 2(R/r)^6)

Seems that Rmin values are written as Rmin/2 in AMBER files avoiding some
calculations to optimize the software. But this occurs at implementation
level and probably does not have correspondence with the paper.

My question is how can I adapt these parameters so they can be used in
Gromacs?

(Sorry about the previous email, now the title is correct)

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3728907/pdf/nihms481931.pdf

[2] ftp://ftp.gromacs.org/pub/manual/manual-5.1.2.pdf

[3] http://comments.gmane.org/gmane.science.biology.gromacs.user/82242


Cheers,

Pedro Lacerda