If you encounter any problems or have any questions in using APBSmem, please be sure to email me at kmc112 (at) pitt.edu or post your question here in this forum. It's very helpful for us to resolve any lingering issues in the software, and we will get back to you as soon as possible. Thanks.
University of Pittsburgh-Carnegie Mellon
I've successfully installed APBSmem in a windows system. However, when I try to run the demos (for instance, protein solvation), it keeps saying "Error: an error ocurred while writing the first dielectric maps".
Pedro L. Loureiro
Thank you for creating that nice application.
I have a few questions about the results and hope you could answer them:
1) Is it possible to load the final EP map on UCSF Chimera? Which is the right file?
2) The membrane representation is tricky. Does the membrane thickness include the lipdheadgroup thickness?
3) In the second example (Ion Solvation) you set the membrane thickness to 24 A., although the value provided in OPM data base (http://opm.phar.umich.edu/protein.php?extrapdb=1bl8) is higher. How did you select that value?
4) Is there any way to apply different titration to the outer, transmembrane and inner part of the protein according to the pH values at each section?
I am looking forward to hearing from you.
1) Yes it is possible to load the electrostatic potential map in UCSF Chimera. It is very easy in VMD and PyMOL, however I find the process confusing in Chimera so here are the step-by-step instructions:
a) When performing your APBSmem calculation make sure you click the checkbox labeled "Draw Potential". This will have APBS write a pot_#.dx file in your results directory.
b) Start UCSF Chimera. Open your PQR or PDB coordinates file.
c) Compute the molecular surface (using MSMS): Actions menu > Surface > Show
d) Open your potential dx file: File > Open.
e) A window titled "Surface Color" should open. In the first drop-down menu, select "MSMS main surface of your_protein.pqr"
f) Click Color.
2) I think I may know what you mean by "tricky" but I am not sure. In APBSmem we represent the membrane by displaying an isocontour of the dielectric map that depends on the membrane dielectric. If you use a different dielectric for the lipid headgroups, then no, you will not see this additional thickness. Since the default is a headgroup dielectric of 80, you could not distinguish between the membrane and solvent, even though there is a big difference in ion accessibility. Perhaps in the future we could use the kappa (ion accessibility) maps to represent the membrane. Until then, it would be easy to do this in VMD, PyMOL or the Volume Viewer in Chimera. You could either set the headgroup dielectric to a value lower than the solvent dielectric or just load the Kappa map.
3) When I go to the OPM link you posted I do not see any data. Maybe you are referring to the coordinate file. In any case, yes, 24 angstroms is quite thin. As discussed in our paper , in this calculation we were comparing our results to a paper by Roux and MacKinnon  and these parameters brought us relatively close to their values without being unrealistic.
4) Unfortunately I am not aware of any software that will do this for membrane proteins. PDB2PQR can be used with PROPKA but that will not account for this. You may need to manually adjust your PQR files if you have some expectations about the side chain charges.
 Callenberg, K.M., Choudhary, O.P., De Forest, G., Gohara, D., Baker, N., Grabe, M. (2010). APBSmem: A graphical interface for electrostatics calculations at the membrane. PLoS ONE 5(9): e12722
 Roux B, MacKinnon R (1999) The cavity and pore helices in the KcsA K+ channel: electrostatic stabilization of monovalent cations. Science 285: 100–102
Hope that helps!
kmc112 (at) pitt.edu
Thank you for your swift response. Below are my comments:
1) Actually it was more of a question about which pot_[1-3].dx file(s) to load in order to reproduce the image created in APBSmem GUI when activating "Draw potential". As far as I know you can't load multiple density maps concurrently in Chimera (I shall ask the developers about it). The reason I prefer that software is because it makes tomographies which may help us understand the translocation mechanism through the EP inside the transporter. But if you know any way to reproduce that image with another software please feel free to tell me.
It would be also helpful to have a section in the documentation describing each output file.
3) Yes, there is not info in the webpage, but there are membrane boundaries inside the structure file which I measured that are separated by more that 24 A.
The different pot_#.dx files correspond to levels of focusing, where the low numbers are the coarse grids and the high numbers are the fine grids. Therefore, for the file with the highest resolution you would probably want pot_3.dx if you used 2 focusing levels, or if you want a larger grid, you could use pot_1.dx.