Previous Thread
Next Thread
Print Thread
AMP-Mg++ charge/LJ parameter determination (NBFIX)
#1148 03/30/04 12:50 PM
Joined: Mar 2004
Posts: 1
D
Forum Member
OP Offline
Forum Member
D
Joined: Mar 2004
Posts: 1
I have developed a topology entry for AMP by analogy with ADP,ATP. 1)Are these partial charges for the final group reasonable:
GROUP
ATOM C5' CN8B -0.08
ATOM H5' HN8 0.09
ATOM H5'' HN8 0.09
ATOM O5' ON2 -0.50 ! charge adj to yield -2.0
ATOM P P 1.10
ATOM O1P ON3 -0.90
ATOM O2P ON3 -0.90
ATOM O3P ON3 -0.90
==========================================================

System: AMP-Mg++ in a periodic box 40A on a side.
5841 atoms
1935 waters
full electrostatics

Goal: Calculate the PMF curve for the dissociation of Mg++ from AMP in solution.

Idea: Create a set of "psf" files with different partial charges corresponding to each window for umbrella sampling.

Method: Use gaussian98 to calculate the partial charges of AMP, Mg++, and 5/6 waters that coordinate with Mg++ at fixed separations between Mg++ and P. Treat the rest of the system as point charges.

Questions: 1)Should we use ESP or Mulliken partial charges? I would think ESP partial charges to reproduce the electrostatic interaction between AMP and Mg++.
2) Since we are trying to explicitly take into account polarization effections, should we rescale certain VdW parameters?

thank you very much,
don burgess

Last edited by alex; 03/31/04 04:26 PM.
Re: AMP-Mg++
deburgess #1149 03/31/04 04:25 PM
Joined: Oct 2003
Posts: 782
A
Forum Member
Offline
Forum Member
A
Joined: Oct 2003
Posts: 782

don,

i would suggest the following charges; they are based on the model
compound mp_2.

GROUP
ATOM C5' CN8B -0.18
ATOM H5' HN8 0.09
ATOM H5'' HN8 0.09
ATOM O5' ON2 -0.40 ! charge adj to yield -2.0
ATOM P P 1.10
ATOM O1P ON3 -0.90
ATOM O2P ON3 -0.90
ATOM O3P ON3 -0.90

For the charges, charmm22/27 charges are based on reproduction of hf/6-31g*
interaction energies. see my web page for more information as well as the
nucleic acid parameter paper.

http://www.pharmacy.umaryland.edu/faculty/amackere/param/force_field_dev.htm

thus, i would calculate the interaction enegy as a function of the Mg++ to phosphate
distance and then adjust the phosphate charges accordingly. Note that the reason
we use interaction energies for the charge determination is that the force field
is designed to accurately treat interactions with the environment rather than
a gas phase charge distribution, which ESP charges give you. In addition, using
ESP charges for part of the system will lead to an imbalance in the nonbond
interactions.

Concerning VDW interactions, yes, you can use them via the use of off-diagnol
LJ parameters. This is done via NBFIX parameters that allow for explicit
Emin and Rmin values to be assigned to an atom pair instead of using the
normal combining rules. See parmfile.doc for more info. Note that in some
earlier releases of the charmm22 protein
parameters NBFIX was used to help treat ions, as you are doing now.

alex






School of Pharmacy
University of Maryland
20 Penn Street
Baltimore, MD, 21201

Moderated by  alex, lennart, rmv 

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.4
(Release build 20200307)
Responsive Width:

PHP: 5.6.33-0+deb8u1 Page Time: 0.011s Queries: 18 (0.005s) Memory: 0.8950 MB (Peak: 0.9594 MB) Data Comp: Off Server Time: 2020-10-01 19:40:02 UTC
Valid HTML 5 and Valid CSS