[gmx-users] re: stacked bilayers and pbc errors

[Senthil Kandasamy]

Answering my own question here:

After a whole lot of trials and errors, I figured out that some of he 
molecules in the initial bilayer were not whole.
So i used trjconv -pbc nojump on the initial bilayer and then used genconf 
-nbox 2 2 1 .
This gave a larger bilayer that in essence did not need energy minimization 
since there are no hard contacts.
I still don't fully understand the way GROMACS treats pbc, but atleast I 
figured this one out.
Without the pbc -nojump option, the stacked  bilayer structure was so bad 
that all the lipid molecules broke into pieces on running md.
It was amusing to see the waters intact, in the original box, but the 
lipids all broken up and exploding. Funniest pdb file I have seen yet.

Senthil