Hi,
Thanks.
> The radius in MEP computation for your metallic center was not defined
> by Singh & Kollman...
No the radius for the atoms were taken from
http://www.ccdc.cam.ac.uk/products/csd/radii/table.php4
and used for the RESP calculation.
>
> Did you optimize the tetrahedral complex by QM? or did you manually
> 'arrange' the molecular system by adding a water molecule?
>
> Did you optimize the trigonal complex by QM as well?
>
> Did you compute RESP or ESP charges?
>
Yes both states were optimized with QM and the basis set should be
fine for copper which is the metal ion in my case. After optimization,
I computed the hessian and from that the force constants.
Also ESP and mulliken were calculated.
> Did you try not using parameters from TIP3P?
> My understanding is that TIP3P is for a solvent...
No - I have not (yet) but I still do not understand how the vdw radius
defined in Amber is neglected during the minimization. I could add a
vdw term for the hydrogen and see if this helps. I have been able to
parametrize a couple of tetrahedral complexes with TIP3P and I have
not experienced any problems there - not during minimization nor MD
simulation.
I will collect some statistics from the server and see if this differs
significantly - that is a good test.
On Thu, Mar 17, 2011 at 3:03 PM, FyD <fyd.q4md-forcefieldtools.org> wrote:
> Jorgen,
>
>> In my crystal structure, I have a tetrahedral coordinated metal site
>> and I would like to observe the changes going to the trigonal state. I
>> have optimized the complex with QM and get the charges using
>> Kollman-Mertz schema as described earlier.
>
> The radius in MEP computation for your metallic center was not defined
> by Singh & Kollman...
>
> How is it treated by the QM code you use?
>
> See http://q4md-forcefieldtools.org/REDS-development/
> http://q4md-forcefieldtools.org/REDS-development/faq.php#17
> http://q4md-forcefieldtools.org/REDS-development/popup/popradius.php
>
>> I use the tutorial from
>> Ross and build the new residues - thanks to FyD here on the list - I
>> have made the residues equivalent the three that coordinates the metal
>> - so I have kept the water molecule which is the fourth coordinating
>> in the tetrahedral complex and I have kept it in the trigonal to get
>> some charge transfer of the metal site - I think by removed it the
>> trigonal structure might be too strongly with respect to binding as
>> well as charges.
>
> Did you optimize the tetrahedral complex by QM? or did you manually
> 'arrange' the molecular system by adding a water molecule?
>
> Did you optimize the trigonal complex by QM as well?
>
> Did you compute RESP or ESP charges?
>
> I would use R.E.D. Server development to generate various sets of
> charges values, and compare them to these you already derived. With
> the statistics module of R.E.D. Server you can now compare RESP, ESP
> and Mulliken charges...
>
>> I have parametrized the amino acids ligand as histidine with a new
>> residue type and the only atom type changes is the coordinating
>> nitrogen atom - charges have been added in the new lib file.
>
> ok
>
>> I build
>> the metal with the water molecule where I have taken the parameters
>> from TIP3P - the types are the same but the charges are different
>> which is reasonable in my case.
>
> Did you try not using parameters from TIP3P?
> My understanding is that TIP3P is for a solvent...
>
>> Then I do a minimization of the
>> complex and the hydrogens go to the negative charges nitrogens on the
>> histidine residue - so the oxygen of the water has a vdw=1.7682 and
>> 0.1512 and the metal it is 1.74 and 1.14 so the repulsion part so be
>> rather strong going below 1.5 A but in the minimized complex the
>> distance is 0.99 between the newly parametrized water and metal and
>> the hydrogens have a distance between the nitrogens of 0.98 A - so
>> somehow the vdw does really work for my parameters or libs - I do not
>> know how to solve this any suggestions or comments appreciated
>
> In the order, I would (i) get various set of charges for your two
> entire complexes optimized by QM, (ii) try to have some ideas for the
> vdW parameters for this water molecule & for the metal, and (iii) (i)
> check the minimization input used.
>
> regards, Francois
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Mar 17 2011 - 08:00:02 PDT
