[gmx-users] How to add a repulsive harmonic potential through define virtual atom?

[masoud aliyar]

 Hello dear GMX users I want to predict the probability binding of ligand
to protein by unguided simulation. But when I increased concentration of
ligand in the simulation box, the ligands has aggregated. So I decide to
add repulsive potential between some atoms of ligands. I had search the
archive of GMX-User mailing list and saw this question: How to add a
repulsive harmonic potential?(1) , and read its answers. Finally according
to the Justin’s Solution I defined one new ATOM in [atomtype] section and
placed it in the [atoms] section of my ligands topology file (lig.itp).
Then I defined the added atom as virtual atom in [virtual_sitesn] section.
Then I create a new topology file (ligand-parameter.itp) that includes
[atomtype] section of ligands (has cut from ligand topology files) and
[nonbond_params] section. Also I added coordination of these new atoms in
my complex coordination file (Conf.pdb). When I run command gmx mdrun -v
-deffnm em got this error: Software inconsistency error: The mass of a
vsiten constructing atom is <= 0. Can anyone help me to fix this error? Log
file of what I do is here (log) (Files are in download links)   1-
https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2015-April/096469.html
Download
links :  http://d01.megashares.com/dl/Jt8d7PZ/lig.itp
http://d01.megashares.com/dl/LguyLcH/ligand-parameter.itp
http://d01.megashares.com/dl/FHhKyVe/conf.pdb
http://d01.megashares.com/dl/aIEDwDN/log