There are no blatant errors. However, the theory that you refer to in the link is an asymptotic approximation to the Poisson-Boltzmann model that APBS uses. There’s no reason to expect agreement between the analytic solution and the numerical solution when they come from different models.
Pacific Northwest National Laboratory
From: Nikolay Rodionov [mailto:email@example.com]
Sent: Monday, September 10, 2012 7:12 AM
Subject: Re: [apbs-users] large potential values
More or less, I am curious if anything blatantly stands out as incorrect in the input file. My solute is a 14nm long particle of the tobacco mosaic virus. There is one thing really sparks my suspicions. When calculating for the surface potential on the external surface, the mean potential is positive at an extremely low molarity and negative at .6M. I don't understand how surface potential can be positive when surface charge density is negative.
The basis of my calculations is eq127 of http://tfy.tkk.fi/kurssit/Tfy-3.363/lectures/lecture06a.pdf ... I solve for charge density in a region by summing the partial charges that make up the surface as according to Solvent Excluded Surface analysis and then dividing by an estimation of the surface area.
On Sunday, September 9, 2012 1:58:51 AM UTC-4, Nikolay Rodionov wrote:
Could you look at my input file and see if there are any errors? I am trying to calculate the ESP of the system at 298K and .6M/10e-7 NaCl solution. The mean surface potential based on APBS is almost exactly an order of magnitude greater than what I get from calculating it by hand using:
Charge_dens = sqrt(8 kB T Co Er Eo) * sinh( (e*zi*SurfPot) / (2 kB T) ), simple solution to PBE with a symmetric ion
....(I can solve for charge density because of the symmetric quaternary protein structure, and then use it to solve for the surface potential)
APBS give me a values of -3.4e3 mV and -5.8e2 mV for the two molarities, while my hand calculations give me -3.9e2 mV and 6.1e1 mV
I've checked my hand calculations and everything is correct so I am guessing that there is something wrong with my APBS input.
Here is my .6M input file:
mol pqr cmplx.pqr
dime 129 129 129
cglen 246.5325050354004 246.43500137329102 264.52200508117676
fglen 246.5325050354004 246.43500137329102 264.52200508117676
cgcent mol 1
fgcent mol 1
ion charge 1 conc .6 radius 1.16
ion charge -1 conc .6 radius 1.67
write pot dx SeaWater-pot
write charge dx SeaWater-chargedist
write atompot flat SeaWater-atompot