[Chimera-users] delphi

Elaine Meng meng at cgl.ucsf.edu
Mon Jan 5 17:28:31 PST 2009


Hi Darren and others,
I forgot to mention a recent development that might interest you:

We recently added a Coulombic Surface Coloring tool (under Tools...  
Surface/Binding Analysis, not yet documented).  This is for people  
who want to color by electrostatic potential but can't or prefer not  
to run any of the continuum electrostatics programs (DelPhi, APBS,  
UHBD) - it associates the atoms with partial charges from AMBER,  
calculates a Coulombic potential, and allows you to color the  
molecular surface by those values.  The Coulombic potential is cruder  
than what you get from Poisson-Boltzmann  calculations, but our  
comparisons on a handful of structures showed the same qualitative  
results (bluest and reddest patches generally in the same places).   
This is only available in the daily builds, not the recent production  
release.  Users should also be aware it is still being improved upon  
and various behaviors may change.

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 Dec 30, 2008, at 4:55 PM, Elaine Meng wrote:

> Hi Darren,
> Unfortunately it is an unsolved problem getting a pre-made delphi
> executable for Mac OSX that gives results that can be shown in
> Chimera.  Just to clarify for everybody, DelPhi is a program freely
> provided by another group, but without official support.
>
> I went through several weeks of trying to get an older executable from
> the Honig site to work, installing various files myself just as you
> have.  Here is the chimera-users message where I describe running up
> against this wall:
> <http://www.cgl.ucsf.edu/pipermail/chimera-users/2006-November/ 
> 001113.html
>>
>
> I reported these problems to the Honig group, and some months later
> (July 2008) they updated their Mac OSX Intel executable so it does not
> require these additional libraries.  Maybe you downloaded an
> executable before that time or have a PPC Mac rather than Intel.
> Honig group DelPhi download page:
> <http://wiki.c2b2.columbia.edu/honiglab_public/index.php/ 
> Software:DelPhi_System_Requirements
>>
>
> However, although the Intel executable runs nicely and can be
> controlled by DelPhiController, something about the map format changed
> and Chimera cannot read the result.  We were unable to get information
> from the Honig lab on how the format changed, or even reverse-engineer
> the changes given our example maps.  There is nothing else we can do -
> it is a limitation recorded in our database as problem report #6228.
>
> Possible solutions are to use DelPhi on a different type of computer,
> use one of the other Poisson-Boltzmann programs that produces files
> that Chimera understands (Grasp, APBS, UHBD):
> <http://www.cgl.ucsf.edu/chimera/current/docs/UsersGuide/ 
> filetypes.html#esp
>>
>
> Actually in May 2008 I found one web server that will calculate an
> electrostatic potential map and let you download it (so results can be
> viewed in Chimera).  There are some complications with using it, but
> perhaps that's still easier than the alternatives.  Here is what I
> wrote in May - haven't tried it since then:
> ------------
> webPIPSA  web server for Protein Interaction Property Similarity
> Analysis
> http://pipsa.eml.org
> focused on performing a calculation on multiple proteins and comparing
> them (won't let you proceed until you specify at least two), choice of
> using UHBD or APBS
> input: upload PDB files or specify PDB IDs, email address for getting
> results
> output: dendrogram and heat map comparing the results for the multiple
> proteins, optional download of zip file with intermediate results
> (includes map files and corresponding PDBs)
>   - my first attempt with two structures and align "none" and use UHBD
> failed...I guess you are only supposed to use "none" if the proteins
> are already superimposed, which they weren't because I didn't care
> about the comparison
>   - second attempt with align option "sup2pdbs" and use UHBD was
> successful.  In Chimera, should probably only view the maps along with
> the corresponding PDBs from the intermediate results since they may
> have been transformed (in the superposition step)
> -----------
>
> Good luck,
> Elaine
> -----
> Elaine C. Meng, Ph.D.                          meng at cgl.ucsf.edu
> UCSF Computer Graphics Lab and Babbitt Lab
> Department of Pharmaceutical Chemistry
> University of California, San Francisco
>                       http://www.cgl.ucsf.edu/home/meng/index.html
>
>
>
> On Dec 30, 2008, at 12:41 PM, Darren Thompson wrote:
>
>> I installed the ibm compiler libraries on Mac OSX tiger, but delphi
>> still does not see them
>>
>> Applications            System Folder           mach.sym
>> Applications (Mac OS 9) TheVolumeSettingsFolder mach_kernel
>> Desktop (Mac OS 9)      Trash                   opt
>> Desktop DB              Users                   private
>> Desktop DF              Volumes                 sbin
>> Desktop Folder          automount               sw
>> Developer               bin                     tmp
>> Documents               cores                   usr
>> Library                 dev                     var
>> Network                 etc
>> System                  mach
>>
>> the opt folder (ibm) is there.
>>
>> I did a symbolic link to the Mac OSX executable in the /usr/local/
>> release1.1/exe directory to /usr/bin/delphi, downloaded the example
>> files from delphi's website,
>> [psb-232-45:release1.1/examples/example_1] darren% delphi param_1
>> dyld: Library not loaded: /opt/ibmcmp/lib/libxlf90.dylib
>>   Referenced from: /usr/bin/delphi
>>   Reason: image not found
>> Trace/BPT trap
>>
>> when I try directly
>> [psb-232-45:local/release1.1/exe] darren% ./delphiMacOSX
>> dyld: Library not loaded: /opt/ibmcmp/lib/libxlf90.dylib
>>   Referenced from: /usr/local/release1.1/exe/./delphiMacOSX
>>   Reason: image not found
>> Trace/BPT trap
>>
>> Does it matter that it's looking for /libxlf90.dylib and I have /
>> libxlf90.A.dylib?
>>
>> Have some xmas charity
>> Help a newb
>>
>> Darren Thompson
>> Graduate Student
>> UCSC
>> 1156 High St.
>> Santa Cruz, CA 95064
>> 831-459-3390  voice
>> 831-459-2395 fax
>> http://people.ucsc.edu/~ministry
>
> _______________________________________________
> Chimera-users mailing list
> Chimera-users at cgl.ucsf.edu
> http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users




More information about the Chimera-users mailing list