VMD-L Mailing List
From: Mayne, Christopher G (cmayne2_at_illinois.edu)
Date: Thu Jun 18 2015 - 14:37:31 CDT
One other note to make...if the amino group is forced to rotate out of plane, e.g., steric clash with an ortho substituent, as the electron pair of the amino decouples from the pi system of the aromatic ring, the geometry is expected to distort more towards tetrahedral. The extreme being an animo group that is perpendicular to the ring system should be tetrahedral. Under these scenarios, the CHARMM force field is unable to represent the changing preference in geometry.
On Jun 18, 2015, at 2:27 PM, Aaron Larsen wrote:
I've measured the deflection of the amino group from a planar geometry and it seems that it's somewhat intermediate between trigonal planar and tetrahedral, which might be a reasonable expectation. I have not compared it statistically to a well parameterized amino-containing nucleobase, however. I have begun paying special attention to the angle and dihedral force constants for the other ring bonds and I believe that if I find cause to strengthen these associations, it may help 'calm down' the aminos. I'll let you know if I make any more progress.
I'm cc'ing VMD-L so that others can benefit from our discussion.
One of my colleagues in our labs has observed this behavior with a molecule that looks similar to guanine. I have tested the geometry optimization setup using ffTK for aniline, 1-naphthylamine, and a substructure of our guanine-like molecule, and I cannot reproduce the distortion towards tetrahedral character for exocyclic amines conjugated to the aromatic system.
Have you made any progress on these molecules?
On Jun 12, 2015, at 10:48 AM, Aaron Larsen wrote:
Of course. The nucleobase geometry before the opt. geometry step was totally planar with respect to the aminos, exactly as you would expect. The output from Gaussian featured the same aminos but now with tetrahedral geometry with respect to the aminos. I corrected the geometry manually in the resulting pdb files, hoping that would be sufficient to continue. However, when I run simulations with the fully parameterized molecule, I continue to see the amino groups adopt a tetrahedral geometry. In fact, the heterocycles seem to distort slightly from a planar geometry overall. This indicates to me that perhaps the torsions or angle force constants are too low.
Any advice on how to correct this issue would be most appreciated.
I'm not sure what you mean by:
> FFTK seems intent on treating these aminos as tetrahedral in geometry
Can you expand on where the perceived problems with geometry are occurring?
On Jun 12, 2015, at 8:03 AM, Aaron Larsen wrote:
> Hello VMD users,
> I'm in the middle of parameterizing a bunch of nucleobase like molecules with the FFTK. Some of them are decorated with aminos. These amino groups SHOULD be planar in geometry, reflecting the aromatic nature of the heterocycles to which they are attached. However, FFTK seems intent on treating these aminos as tetrahedral in geometry. It seems that there should be a simple work around for this. Would anyone have any suggestions on how to proceed?
> It would be preferable if the solution did not involve starting over from the Molefacture step.
-- Aaron Larsen, Ph.D. Harvard University Department of Chemistry and Chemical Biology Harvard Medical School Department of Genetics E-mail: alarsen_at_molbio.mgh.harvard.edu<mailto:alarsen_at_molbio.mgh.harvard.edu> Mobile: 617-319-3782<tel:617-319-3782> FAX: 617-643-3328<tel:617-643-3328> -- Aaron Larsen, Ph.D. Harvard University Department of Chemistry and Chemical Biology Harvard Medical School Department of Genetics E-mail: alarsen_at_molbio.mgh.harvard.edu<mailto:alarsen_at_molbio.mgh.harvard.edu> Mobile: 617-319-3782 FAX: 617-643-3328