[Chimera-users] huge electrostatic range for APBS

Elaine Meng meng at cgl.ucsf.edu
Thu Aug 7 09:40:38 PDT 2014


Hi Dennis,
Make sure that the range is calculated from the surface that goes with the same set of coordinates as used to calculate the map, because another reason you could get a large value is that the surface is based on a different set of coordinates, and thus comes very close to some atom in those coordinates.  Also make sure that the map hasn't been displaced from the coordinates/surface model(e.g. matrixcopy #0 #1 if the current coordinates are #0 and the map is #1), although if the coloring looks OK it is unlikely to be the problem.

To further investigate the ESP map:
You could use Volume Viewer to show a histogram of the grid values (not the projected-surface values), and show isopotential contours to help identify the region(s) of large magnitude.  If you already have the map open, just start Volume Viewer (in menu under Tools… Volume Data) and click the little oval "eye" outline to show the histogram and isopotential surface sliders.

Waters/ions probably aren't the issue if these structures contain the same sets of atoms, just at different points from the simulation.  However, we added an option to the APBS interface to exclude solvent (default true) as of Chimera 1.9 because people were accidentally including them in the APBS calculation.  It was easy to forget about the waters because they are generally undisplayed.

I hope this helps,
Elaine
----------
Elaine C. Meng, Ph.D. 
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco



On Aug 6, 2014, at 8:19 PM, Dennis N Bromley <dbromley at uw.edu> wrote:

> thank you Tom, this is great.  It gives me something to work with.  Another part of my question is why two structures of the same protein would give such dramatically different ranges.  Is it again due to some local structural feature where one protein has a surface that allows close access to a charged atom and the other doesn't?  But that doesn't make much physical sense unless the surface calculation is doing something funny...?
> 
> 
> On Wed, Aug 6, 2014 at 6:39 PM, Tom Goddard <goddard at sonic.net> wrote:
> Hi Dennis,
> 
>   The “Set full range” button when coloring the molecular surface using APBS electrostatic potential values is looking at the minimum and maximum potential values on the surface (actually 1.4 Angstroms out from the surface).  Your question is why do you observe such high values with the APBS potential, while the Chimera Coulomb potential never seems to give high values.  The high potential values are at points very close to charged atoms.  Why would that happen in the APBS case?  One thought is that the APBS calculation included water or ion charges while the Coulomb calculation did not, and those ions were close to the surface and the 1.4 Angstrom offset from the surface ended up measuring the potential right on top of an ion.
> 
>   In any case using the “set full range” button is not very sensible in this situation because it sets the coloring range using the most extreme values which are outliers.  But it is good you tried it since it seems to suggest something is perhaps wrong with APBS calculation, or at least somehow potential values are being used at points that are unexpectedly close to charged atoms.
> 
>         Tom
> 
> 
> On Aug 6, 2014, at 3:13 PM, Dennis N Bromley  wrote:
> 
> > Hi all,
> >
> > I have two proteins models from different points in an MD simulation.  I colored the surface by Coulombic Surface Coloring, which ran AnteChamber and such.  Then I ran APBS.  When I color by the APBS data, they look more or less similar.  But when I click the "Set full range" button, the kBTe unit range on one is about +-3 and the unit range on the other is about +-300.  Any ideas why one would be so much larger?  Or am I misinterpreting something?  Both were run with exactly the same parameters, all default except for temperature = 310.
> >
> > thanks-
> > -denny-
> >




More information about the Chimera-users mailing list