The Molecular Graphics Laboratory Forum

Hi All,

So, I've just started using these tools and I apologize if this is a well understood issue. I searched the forums but didn't find anything.

I'm looking at virtual screening and grabbed a ZINC package to putter around with. When I ran prepare_ligand4.py against the ZINC ligand (ZINC05759399) I get this output and error:


--------------stderr output----------------------
Traceback (most recent call last):
File "c:\Program Files (x86)\MGLTools 1.5.4\MGLToolsPckgs\AutoDockTools\Utilities24\prepare_ligand4.py", line 207, in <module>
if a.coords!=coord_dict[a]: bad_list.append(a)
KeyError: <Atom instance> tmpFFA0.tmp:default:<0>:H1
----------------------------------------------------

-----------stdout output--------------------------
set verbose to True
set ligand_filename to C:\Users\dbromley\AppData\Local\Temp\1\tmpFFA0.tmp.mol2
set outputfilename to C:\Users\dbromley\Desktop\ligandTest\ZINC05759399.pdbqt
read C:\Users\dbromley\AppData\Local\Temp\1\tmpFFA0.tmp.mol2
setting up LPO with mode= automatic and outputfilename= C:\Users\dbromley\Desktop\ligandTest\ZINC05759399.pdbqt
and check_for_fragments= False
and bonds_to_inactivate=
returning 0
-------------------------------------------------------

It still produces what appears to be a well-formed .pdbqt file. Can I ignore this? What does this mean?

FYI, other ZINC ligands (e.g. ZINC01752827) do not cause this error so I don't think it's a systemic problem in my screening pipeline as such unless I'm tickling some corner case.

thanks for any help!
-denny bromley-

Thank you for the message. Could you please upload mol2 causing this problem as an attachment for this topic? Thank you.

_________________
Sarkis Dallakian - Google Profile - PyRx - Nutrition Facts | Food Prints - Clinical Trials Finder | ClisMap

thanks for the reply - here you go.
-denny-

I have confirmed that I get the same error with the .mol2 file downloaded directly from the zinc website.
http://zinc8.docking.org/srchdb.pl?zinc=5759399

hmmm.... Hopefully I'm just missing something.

thanks!
-denny-

Thanks for the link and attachment. I've run the following command and it did not give any error messages:


Could you please post the command you use to call prepare_ligand4.py? If your have earlier version (< 1.5) of MGLTools installed, uninstalled it and make sure that there is no MolKit or AutoDockTools folder in C:\Python25\Lib\site-packages. Also, see if your pdbqt is different from the attached pbdqt. Thanks!

_________________
Sarkis Dallakian - Google Profile - PyRx - Nutrition Facts | Food Prints - Clinical Trials Finder | ClisMap

Hi, I have the same error on many .pdb files.

The error appears only when using the option -A (both bonds_hydrogens or hydrogens).

The MGL version is 1.5.4, and I use Mac OS X 10.6.2.

The problem seems to be located at line 206 of prepare_ligand4.py,
when the script try to tell the user the number of atom coordinate changes:

the dictionary "coord_dict" doesn't contain the key "a".

I've changed a value in the .PDB file: in the column number five, all values was = 0.
I set value = 9 for the atom I think there was some problem using the add
hydrogens parameter.

In addition, I've added a line in the prepare_ligand4.py.

First was:

for a in mol.allAtoms:
if a.coords!=coord_dict[a]: bad_list.append(a)

Now is:

for a in mol.allAtoms:
print a
if a.coords!=coord_dict[a]: bad_list.append(a)


The print command allow me to see:

<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK0:P
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:C
<Atom instance> labm.mlims-10070: :UNK0:O
<Atom instance> labm.mlims-10070: :UNK9:N
<Atom instance> labm.mlims-10070: :UNK9:HN1
Traceback (most recent call last):
File "/Library/MGLTools/1.5.4/MGLToolsPckgs/AutoDockTools/Utilities24/prepare_ligand4.py", line 208, in <module>
if a.coords!=coord_dict[a]: bad_list.append(a)
KeyError: <Atom instance> labm.mlims-10070: :UNK9:HN1


Please note the lines:
<Atom instance> labm.mlims-10070: :UNK9:N
<Atom instance> labm.mlims-10070: :UNK9:HN1

We see UNK9 and not UNK0!
So I think that the problem becomes by adding hydrogens relatively to the HETATM line number 13 in PDB:

Originally was:
HETATM 13 N UNK 0 5.816 7.361 1.202 0.00 0.00 N+0

And for this "experiment" now is:
HETATM 13 N UNK 9 5.816 7.361 1.202 0.00 0.00 N+0

I don't know the code, so my qyestion is: I've to pay attention to this error (and don't use the .pdbqt
file for docking) or it's only a little bug that doesn't change anything of really important?

Please tell me something and many thanks to you and all the Staff.

The (original) pdb file is in attachment, and if may hope I've many others pdb files that generate the same error.
Olso, the generated pdbqt is in attachment.

Bye
Alberto

Sorry to revive an old thread but I'm having this problem with a different ligand (hydroxytamoxifen OHT, extracted from the ERalpha/OHT complex structure on PDB). I am using pyMolMac (X11 hybrid) in combination with Daniel Seeliger's autodock/vina plugin for pyMol. As others have said, the pdbqt file is produced anyway (but the error message messes with the plugin which is very irritating -- it thinks the ligand hasn't been generated). The error is this:


The relevant area of the script is this:



where I have commented the two offending lines. The weird thing is that there is no error message of the type "xxx atom coordinates changed!" as there should be if bad_list is nonzero length (I don't know Python but I'm pretty sure this is right).

I don't really have any idea other than that when those lines are removed it works fine. Can anyone help me out? Does removing the lines present any potential for erroneous ligands to be generated? Please let me know if I need to provide more info.

Cheers,
Jack