Haddock: error uploding a PDB file with metals

HADDOCK
1

Hello Dr. Bonvin and All,

I’m having some problems regarding the docking between two proteins tin order to make a homodimer. Both of proteins are in their holo state, with a Zn ion and a Cu one.
At first, when I tried to upload them into Haddock, it returned me the following error:

“Error in PDB file.
Issue when parsing the PDB file at line 1139 .
Your PDB contains an elemental ion ZN without specified charge”

In order to solve it, I changed the PDB following the guidelines at the following link: HADDOCK Web Server - Settings

In particular, I wrote ZN+2 instead of ZN for the atom and ZN2 instead of ZN for the residue.

When I uploaded the new PDBs into haddock, it returned me the following error:

“Error in PDB file.
Issue when parsing the PDB file at line 1139 .
Your PDB contains multiple forms of the same residue ZN2 201 . This is not supported in the current form. If you would like to supply multiple conformations, please create an ensemble.”

After that error, I read a bit on the forum and I tried to use the web pdbtools, uploading the structure and running the code just using the main: “pdb_selaltloc”, without pre-processing or post-processing operations.

So I uploaded the third new PDB file, the one from pdbtools, into Haddock, and it returned me the following error:

“Error in PDB file. Unknown error: invalid literal for int() with base 10: '+ '”

I searched this last error on the forum and I didn’t came up with a solution yet.

In the following image, there are the last parts of the PDB files used when I encountered these errors:

  • The original one;
  • The corrected one with the guidelines from Haddock library;
  • The corrected one using the pdb_selaltloc tool;

image
image762×508 82.6 KB

Is it possible to help me? I hope I sent you enough information regarding the problem, if not I will surely remedy.

Thank you for the attention, and keep up with the good work!
Cordially
Marco

2

You are missing the 2 next to the + in the atom name

1 Like
3

Thank you for the fast response Dr. Bovin

I just did it, modifying the last PDB:

image
image964×129 3.94 KB

(Also the columns seem to be not aligned after the use of the pdb_tool)

It returned me this error, though:

“Error in PDB file.
Issue when parsing the PDB file at line 1111 .
ATOM/HETATM line does not meet the expected format”

4

The PDB format is quite strict. By adding the two you shifted all other columns (clearly visible from the snippet you sent)

Delete one space after the 2 you added.

1 Like
5

Yeah, it seems to work now, thank you very much.
One last thing:
Is it normal that the last line:
TER 1111 GLN A 153
has been swapped with the following one:
TER 1113 CU2 A 202 ?
Also the conects are gone.

image
image756×289 89.7 KB

6

CONECT are irrelevant for HADDOCK

And don’t worry about the TER lines. This is a correct behaviour

1 Like
7

Thank you very much for your help, Dr. Bonvin!!

8

Hello,
is it possible that the pdb_selaltloc tool change ions charges? I’m comparing the pdb structures before and after having used the tool, and the zinc ion charge changed from +2 to +1, even if the copper ion charge remained +2. I noticed that using MOE software, protein->properties->esemble residue charge

Tha’s the pdb before:
ATOM 1108 OE1 GLN A 153 50.819 4.796 5.554 1.00 27.09 O
ATOM 1109 NE2 GLN A 153 50.228 2.700 4.908 1.00 33.93 N
ATOM 1110 OXT GLN A 153 50.916 2.972 10.999 1.00 22.53 O1-
TER 1111 GLN A 153
HETATM 1112 ZN+2ZN2 A 201 38.320 16.114 32.936 1.00 18.06 ZN2+
HETATM 1113 CU+2CU2 A 202 40.908 19.845 28.000 1.00 23.14 CU2+
CONECT 420 1064
CONECT 1064 420
END

Tha’s the one after selaltloc
ATOM 1108 OE1 GLN A 153 50.819 4.796 5.554 1.00 27.09 O
ATOM 1109 NE2 GLN A 153 50.228 2.700 4.908 1.00 33.93 N
ATOM 1110 OXT GLN A 153 50.916 2.972 10.999 1.00 22.53 O
HETATM 1111 ZN+ ZN2 A 201 38.320 16.114 32.936 1.00 18.06 ZN
HETATM 1112 CU+ CU2 A 202 40.908 19.845 28.000 1.00 23.14 CU
TER 1113 CU2 A 202
END

I don’t notice difference or informations about charge reading those pdb lines (In fact I used MOE to do that).
if that’s normal, do you know how to manually recover the +2 zinc charge, modifying the pdb?

Thank you very much
Have a nice day
Marco

9

Looks like a “feature” to me …

You could open an issue in the pdb-tools git repo

10

PS: indeed simple editing can fix this