[gmx-users] Net charge neutrality during thermodynamic integration

[Leandro Bortot]

Dear users,

     I'm learning how to use the Thermodynamic Integration approach to
calculate free energy differences. I have done some tutorials, including
those available at www.alchemistry.org.
     however, all tutorials I found use neutral ligands for the decoupling.
Since I use cutoffs with PME and some of my ligands are charged, I wonder
how should I treat the fact that at labda=1.0 the system net charge will be
neutral and at lambda=0.0 it will be a non-zero integer value because the
ligand charges has been set to zero.
    should I decouple a counter-ion together with the ligand or should I
let the charge be non neutral and apply some sort of corrections in the
results?


     If possible, could you provide some paper or tutorial in which this is
done and explained? Although I could find some papers in which charged
ligands are decoupled, no explanation was given about the net charge.


best regards,
Leandro.