[Chimera-users] ESP surface, DelphiController

[Elaine Meng]

Hi James,
Just because it has red, white and blue blotches does not mean it is  
showing ESP!!

I don't think the different filenames in the DelPhi log are a problem  
- I believe temp files are generated from the ones you specify and  
used in the calculation for technical reasons.

However, I do not know the reason for the error and not getting any  
output.  You could try using Help... Report a Bug in the Chimera menu  
- give more details, such as all the error messages obtained and what  
it says at the end of the DelPhi log file.

The main difficulty with DelPhi and any ESP calculation program is  
that atom names need to match those in the DelPhi charge parameter  
files.  If you want to use all the hydrogens, you have to either edit  
all the names in the DelPhi charge parameter file or write out your  
PDB beforehand and edit all of its atom names to match those in the  
DelPhi charge parameter file.  I have found this to be well nigh  
impossible.  In general many hydrogen names not match and your  
molecule gets a very large net negative charge.  Chimera tries to name  
hydrogens according to current PDB standards, but these old DelPhi  
parameter files use a different set of names.

Instead, I would just used the DelPhi parameter files "default.crg"  
and "default.siz" which ignore hydrogens and just plop positives on  
nonhydrogen atoms in Lys, Arg (maybe His and N-term, I forget) and  
negatives on nonhydrogen atoms in Asp, Glu (maybe C-term).

It is impossible to tell what "half all red" and "half all blue"  
means.  It depends on what levels you have specified to be displayed  
as red and blue.  Maybe the APBS map is fine - try changing the values  
set to red and blue to larger magnitudes.  If APBS writes some sort of  
log file I would look at that, see if there is a reasonable net  
charge.  Again most of the trickiness is in getting the right charges  
assigned to the atoms.  You could use the PDB2PQR server to get an  
input file for APBS (if you didn't try that already).
<http://nbcr.sdsc.edu/pdb2pqr/>

Another possibility is to use the webPIPSA server as mentioned in  
previous messages, e.g.
http://www.cgl.ucsf.edu/pipermail/chimera-users/2008-August/003006.html
Best,
Elaine
-----
Elaine C. Meng, Ph.D.                          meng at cgl.ucsf.edu
UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab
Department of Pharmaceutical Chemistry
University of California, San Francisco
                      http://www.cgl.ucsf.edu/home/meng/index.html



On Aug 27, 2008, at 7:58 PM, James Fethiere wrote:

> Hi Elaine,
>
> Nice that you found what I did 6kkm away!!! thanks.  However, I must  
> say that it does look artificial.  I'm just desperate to get the  
> electrostatic look as nice as possible (just like the pictures in http://www.cgl.ucsf.edu/chimera/tutorials/maps08/exercises.html#esp) 
>  and found that was the closest I could get.  One example I'm trying  
> is on 1DH3
>
> On this note, I did try the delphicontroller to get the true  
> potentials but cannot get it to work.  I have installed delphi on  
> windows-XP and called it from within chimera with tool/ 
> surfaceanalysis/delphicontroller. I entered the path for the  
> executable and charge and radii files. When I run delphi, I get an  
> error with no output file generated and some strange parameters in  
> the log file:
>
> in(pdb, file="tmph4vkzc")  This does not correspond to the chimera  
> model I have chosen
> in(siz, file="tmplk_go0")  same here; not the file specified
> in(crg, file="tmpbg7nlz")  same thing
> AC = true
> CS = true
> PF = 60
> SC = 1.2
> PR = 1.4
> FC = true
> LG = false
> LP = false
> PX = false
> PY = false
> PZ = false
> BC = 4
> ID = 2
> ED = 80
> GC = 0.0
> IS = 0.0
> IR = 0.0
> NI = 0
>
> P.S.  I did addH before starting Delphi! Any idea of what is going  
> wrong?
>
> I also tried to run APBS and got a potential file. But when I read  
> it in chimera, I get strange looking surfaces with half all red and  
> half all blue!!! here again I could use some guidance.
>
> Thanks for your help.
>
> James
> -- 
> James Fethiere, Ph.D.
> Institut de Recherche en Immunologie et Cancerologie
> Pavillon Marcelle Coutu, University de Montreal
> 2900 edouard-montpetit
> Montreal, Qc  H3t 1J4
> Tel: 514-343-6111 ext. 0918/0919
> Fax: 514-343-5839
> james.fethiere at umontreal.ca