ERROR IN LigParGen output: wrong atom partial charges - GLycerol molecule

[paolobotto]

Further to my previous issue on Glycerol MD simulation, I would like to inform you about what follows.

1. OPLS-AA FF Atom partial charges from LigParGen Output:

C1 | -0.0315 C2 | 0.0187 C3 | -0.0215 O4 | -0.5871 H5 | 0.1071 H6 | 0.1071 H7 | 0.4237 O8 | -0.592 H9 | 0.147 H10 | 0.4303 O11 | -0.6052 H12 | 0.0906 H13 | 0.0906 H14 | 0.4221

2. OPLS-AA FF Atom partial charges from Jahn, Akinkunmi, Giovambattista, Journal of Physical Chemistry, 2014, Supplementary Information

C1 | 0.145 C2 | 0.205 C3 | 0.145 O4 | -0.730 H5 | 0.465 H6 | 0.060 H7 | 0.060 O8 | -0.730 H9 | 0.465 H10 | 0.060 O11 | -0.730 H12 | 0.465 H13 | 0.060 H14 | 0.060

The differences are outstanding. Lammps simulation of GLycerol molecule works fine with data from 2). LigParGen generated partial charges - 1) - when applied cause the molecule to blow away after some hundreds fs.

I would be greatly interested in receiving some explanation abut this surprising error, of course if somebody is reading this message.