[gmx-users] RE: ligand parameterization for amber port in gmx

[servaas michielssens]

What do you mean by severe deformations? I would expect a lagere
difference in conformation for pyrophosphate bound to mg and free
pyrophosphate. You could do extra quantum calculations and compare them
to your force field calculations to check the quality of your force
field. If this is no good, you could consider using RESP charges. Did
you compare your parameters do published pyrophosphate parameters e.g.
http://www.pharmacy.manchester.ac.uk/bryce/amber

    hope this helps,
    
    servaas


> Dear people,
> 
> I have parameterized a ligand with one phosphate and one pyrophospate group
> using antechamber with AM1-BCC charges and the GAFF forcefield. Amber files
> were converted to gmx files (*.itp/*.top and *.gro) with the amb2gmx
> conversion tool (http://www.alchemistry.org/wiki/index.php/Free_Energy_Tools)
> and then used in gromacs for energy minimization. The geometry looks fine
> after in vacuo EM and EM in water (ffamber tip3p model from amber ports) of
> the ligand alone (not bound to protein!). Minimizing the ligand in its
> binding mode as seen in the corresponding PDB (using the amber03 ff) also
> works fine if magnesium ions are NOT included. Including magnesium ions
> however leads to severe deformations of the phosphate and pyrophosphate
> groups. I suspect it has something to do with not including the magnesium
> ions in the parameterization. Can anybody please give me a hint on how to
> solve this problem?
> 
> Thanks a lot,
> Merc