From: Chris Harrison (char_at_ks.uiuc.edu)
Date: Fri Feb 13 2009 - 17:57:21 CST
To answer your questions:
1) If this concept is correct, should I do the simulation in solvated
systems, with PBC, PME?
This is generally true, ASSUMING sufficient sampling of MUT and WT such that
contributions to their free energies from their conformational free energy
differences are negligible. If MUT's ensemble exhibits a significantly
enough changed ensemble of conformations from that of WT, then MUT and WT
must each be sampled enough in the other's conformational space, to account
for conformational free differences contributing to the overall free
change. This is why perturbations larger than say Tryptophan generally
require extensive, thorough sampling.
Yes, you should use a solvated system with PBC and PME. Long range
electrostatics can have a significant effect, particularly when the
perturbed set of atoms exhibit large columbic/electrostatic interactions
with the rest of the system, as opposed to largely VDW interactions.
2) since the positive charge of the arginine disappears, I guess I need a
ion to appear in lambda=1?
In the purest sense, yes. But this is generally not done and results in
reasonably dG or dA estimates. Further explanation can be provided if
3) is it possible to evaluate the individual contributions to the free
energy difference? (electrostatic, VdW)
These individual energetic "components" are listed in the fep output file
that will be produced when using NAMD for an fep calculation. You will be
able to analyze these as you see fit.
NOTE: if you are using a pre-compiled NAMD2.6, I recommend you download the
CVS version of NAMD as it includes several improvements to free energy
calculations, yielding better results, dG values, etc.
-- Chris Harrison, Ph.D. Theoretical and Computational Biophysics Group NIH Resource for Macromolecular Modeling and Bioinformatics Beckman Institute for Advanced Science and Technology University of Illinois, 405 N. Mathews Ave., Urbana, IL 61801 char_at_ks.uiuc.edu Voice: 217-244-1733 http://www.ks.uiuc.edu/~char Fax: 217-244-6078 On Fri, Feb 13, 2009 at 4:43 PM, Luis Cunha <luis.cunha_at_mssm.edu> wrote: > Dear NAMD users, > I need your advice. I use NAMD to study ligand binding to my protein, but > so far I've only been interested in conformation changes, not energies. I > need advice with the following problem: > I have an xray structure of a protein (actually a domain of the > protein)/DNA complex. In this complex, an arginine makes a Hbond with a > specific DNA base. We've found patients with a specific genetic disease who > have a mutation (Arg to Cys) of this very arginine. Obviously this mutation > has profound effects in the way the protein interacts with the DNA, hence it > causes a disease. I'm hypothesizing that the Hbond removal in the mutant > severely affects the binding capacity. An assay is being developed by > someone else to test this in vitro. Meanwhile, I would like to estimate the > difference in the binding energy between the mutant and wild type proteins. > > Please let me know if I got this correctly: What I need is a thermodynamic > cycle: > > deltaG1 > WT-DNA ---------> WT + DNA > | deltaG4 | deltaG2 > | | > MUT-DNA --------> MUT + DNA > DeltaG3 > And the deltadeltaG (the difference in free energy of binding between the > mutant and the wild type) that I'm interested in would be given by > subtracting deltaG4 from DeltaG2. These 2 deltaGs would be calculated in > two alchemical FEP/MD experiments. > > 1) If this concept is correct, should I do the simulation in solvated > systems, with PBC, PME? > 2) since the positive charge of the arginine disappears, I guess I need a > ion to appear in lambda=1? > 3) is it possible to evaluate the individual contributions to the free > energy difference? (electrostatic, VdW) > > > thank you in advance to any help provided, > > > Luis >
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:21:44 CST