[gmx-users] Xavier Periole,your mean?

[Xavier Periole]

What I mean is that :
1: I don't think your minimization has to be redone with
a different protocol. Especially if the conformation has
already minimized by Peter.
2: you have very strange input for your simulation. What
I think is stange is :
    - your integration time step (dt) is 0.0001ps=0.1fs
    which 10 times smaller than the 1fs (or 2fs with lincs).
    That should not be a problem but it seems to me that it
    if you don't have a specific reason, it is useless. You
    just increase the duration of your simulation by a
    factor of 10.

    - you don't control the pressure.
    - you use position restraints : any reason for that ???
    - you use LINCS. Some times LINCS and position restraint
    don't like each other and LINCS does not converge.

So change what you think is necessary to change in your
input. And rerun your MD.
I would change the dt, take out the position restraints,
put some pressure control and see if it runs.


Good luck 
Xavier


you should concentrated in why