These parameters appeared in TYR.st file, in the folder of examples\myoglobin\ of MEAD suite. Most other parameters can be found in Amber forcefield data sets (An All Atom Force Field for Simulations of Proteins and Nucleic Acids, 1986). I saw partial charge parameters of protonated TYR. But MEAD also requires deprotonated TYR. How to deal with this?
Article Electrostatic Calculations of sidechain pKa values in myoglo...
We did these calculations 20+ years ago, and I can't completely figure out where we got the parameters, but I think they're based on methoxide. The TYR.st files for the "CHARMM" and "OPLS" charges give charges -0.85 and -0.15 for O and CZ, respectively. These are very similar to the methoxide parameters listed in Lim, Bashford and Karplus, JPC 95:5610 (1991). For the "Amber" charges in the myoglobin TYR.st files, the charges are -0.85 and +0.118. This is probably the result of choosing -0.85 for O, and then and then setting the CZ charges so as to make the total charge come out right (prot chargess - deprot charges = 1).
These are not particularly good charges for TYR-. I think that if one did a good QM calculation the negative charge would delocalize more. But in a pKa calculation of the type done here, the highest electric moment dominates, so it's probably not too bad. particularly in a paper where the main focus is on the histidines.
By the way, this paper was the first to present Poisson-Bolzmann type calculations for protein sidechain titration that also included His tautomerization.
Well, since there is no clear definition of the volume, shape and size of an atom/ion in a solid./molecule, there is no clear definition of charge for an ion. However, the Bader's charge analyse approach ([e.g. R. F. W. Bader, T. T. Nguyen and Y. al, Prog. Phys. 44 (1981) 893] provides an unique way to obtain charges at different ztomic sites. Recently, using the first-principles DFT (code VASP) I tried H2O molecule, and ice, water, and obtained charges: H + 0.64, O -1.28 for an H2O molecule. There are some changes in water and ice.
In short, you may try to do some DFT calculations to obtain charges for the atoms/ions you want to use in atomistic approaches.
We did these calculations 20+ years ago, and I can't completely figure out where we got the parameters, but I think they're based on methoxide. The TYR.st files for the "CHARMM" and "OPLS" charges give charges -0.85 and -0.15 for O and CZ, respectively. These are very similar to the methoxide parameters listed in Lim, Bashford and Karplus, JPC 95:5610 (1991). For the "Amber" charges in the myoglobin TYR.st files, the charges are -0.85 and +0.118. This is probably the result of choosing -0.85 for O, and then and then setting the CZ charges so as to make the total charge come out right (prot chargess - deprot charges = 1).
These are not particularly good charges for TYR-. I think that if one did a good QM calculation the negative charge would delocalize more. But in a pKa calculation of the type done here, the highest electric moment dominates, so it's probably not too bad. particularly in a paper where the main focus is on the histidines.
By the way, this paper was the first to present Poisson-Bolzmann type calculations for protein sidechain titration that also included His tautomerization.
I agree.
In most QM calculattions, there are still parameters, such as Wigner-Seitz spheres, to determine the shapes of atoms. Therefore, the obtained electrons in an stomic sphere depend on the settings.
Relatively, using the electron density distribution in a solid/molecule, Bader separated an atom well from the surrounding atoms with a clear definition. Therefore, the obtained charge at one atom/ion site is uniquely determined. (See Bader papers)
Thank you all the replies! I am obtuse enough that I did not notice the question was answered a week ago. I thought I would be notified when the question was answered.
To Donard, I should appreciate this article of yours. It really helps on knowing details of using MEAD. I am trying Hv1 and going to try eGST. In both case, the interesting sites include HIS, while TYR or ARG not probably deprotonated.
To Changming, thank you for the advices. I will read the paper carefully.
Anyway, the choice of partial charge is somewhat arbitrary, so long as the parameters set is self-consistent. CHARMM can also be tried. Am I right?
Dear Xin-Yu,
You are right.
The Bader's charge analysis can be used as references.
Dear Changming, Thank you for introducing me such an important theory that I never heard of! I think it is a powerful tool in many aspects. But perhaps it is beyond my intelligence to use it. Is there any ready-made force fields based on Bader's charge analysis?
Dear Xin-Yu,
We used the Bader charge model to build potentials. Please see the paper:
A transferable force field for CdS-CdSe-PbS-PbSe solid systems
THE JOURNAL OF CHEMICAL PHYSICS · DECEMBER 2014
(You can get the reprint in my researchgate)
Best regards
changming
@Changming, great job! Does the force field can be applied when considering interaction with water or organic compounds such as protein?
- Badges
- Science topic
- Similar topics
- Chemistry
- Theoretical Chemistry
- Computational Chemistry
- Molecular Dynamics