[Dock-fans] ERROR while runing the demo and other issues

Terry Lang terry at cgl.ucsf.edu
Fri Jul 15 19:03:41 PDT 2005


Dear Amit,

         You first issue is that you are using the sphere file that came 
directly out of the sphgen program.  In the current release (5.2.0), 
this would create the segmentation fault that you were seeing.  It turns 
out that DOCK was not properly handling files with more than one 
cluster.  We have resolved this issue and posted a bug fix (#9) at 
http://dock.compbio.ucsf.edu/DOCK_5/DOCK_bugfixes.htm under 5.2.0 bug 
fixes.  There are instructions on this page for how to apply bug fixes 
in addition to the bug fixes themselves.  If you do not want to apply 
the bug fix, you can simply remove the other sphere clusters from your 
10.sph file.         
            I am not convinced that this completely resolves your 
problem, however.  Please test one of the above fixes and let me know if 
you are still having problems.  If you are, you will need to send me all 
of your input files for your DOCK run so that I may do more 
diagnostics.  Simply tar and gzip them and attach them to an email.  
Please send the files only to me so that you don't overload the 
discussion server.

Good luck,
Terry


Amit wrote:

Dear Terry,

First of all - thanks.
Second, I attached a sample ligand (4.mol2), a peptide with a single 
disulfide and also another file (4fixed_amber.mol2), which is the same 
peptide but with a broken disulfide. The later-ligand (4fixed ....) 
docking went fine while the first-ligand (4.mol2) docking caused an 
immediate "segmentation fault". [The same phenomena have occurred with 
less-wired-looking peptides :-)] Also attached is the grid.nrg (hope its 
not too big for your mailbox) and my sphere file named "10.sph". The 
dock.in is "4fixed_amber.in". With both peptides I have used the AMBER 
force field and the Gasteiger charge definitions.

Many many thanks for your time and help,
Amit

Terry Lang wrote:

Dear Amit,

         For your first issue, if you are generating chemical grids with 
the program "grid," you need to remove the entry
______________________________________________________________________
name        null
definition    *

from the chem.defn file.  This entry in the parameter file is not 
backwards compatible with the "grid" program.  That being said, there is 
no function in "dock5" that reads in chemical grids.  Chemical scoring 
did not enhance success rates and we have decided not to reimplement it 
in the new version of DOCK.   We will be removing that capability from 
the "grid" program in the next release to remove confusion.          For 
your ligands, I am not sure what the problem is exactly.  Could you send 
a copy of your ligand files and your dock input files?  I can take a 
look and see if I can figure out what is going on.

Thanks,
Terry

-- 
P. Therese Lang
Kuntz and James Labs
UCSF--Chemistry and Chemical Biology
Phone: (415)476-3986

 





More information about the Dock-fans mailing list