ADL: Attempting to use AutoDockTools with protein structure containing Zn: It crashes...My mistake or a bug?

Ruth Huey rhuey at
Tue Jan 8 06:49:13 PST 2008

David wrote:
> Hi:
> I'm trying to use AutoDock Tools to set up a docking run. My receptor 
> protein has a Zn ion in it.
> If I try to read in the pdb file through AutoDock tools (File->Read 
> Molecule) and then attempt
> to add hydrogens, I get a python error about
> ValueError: Could not find atomic number for Z Z
> Zn is in the parameters file used for autogrid and autodock, but doesn't 
> seem to be in the
> parameters list that is inherently used in ADT, and I can't figure out 
> how to fix that (which
> may mean, as the title says, I'm an idiot).
> If I try to ignore the error, things just get worse (it won't add 
> hydrogens, so flexible residue
> assingments/assigning charges/etc. don't work).
> Is there some trick to using Zn?
> (emails appreciated, since I don't subscribe to this list).
> Thanks,
> David
> ________________________________________________
> --- ADL: AutoDock List  --- ---
Your problem probably is caused by an improper input file:
Specifically, the pdb standard requires that the names of atoms with 
two-character elements start in column 12 (zero-based)  and those of 
one-character element in column 13. [I cannot be certain this is the 
issue without looking at your file]

Here is a snippet of a pdb file showing a properly formatted ZN. Note 
that the hydrogen H is in column 13 while the zinc Z is in column 12.  
The pdb parser in ADT interprets a ZN starting in column 13 as an atom 
of element "Z" (for which there is no information)

ATOM   4260  H1  ASN   285      50.523  49.273  69.914  0.00  
0.00           H
ATOM   4265 ZN   ZN    286      38.027  48.584  69.337  0.00  
0.00          ZN


More information about the autodock mailing list