File problems: docking a short polymer to a protein

HADDOCK
1

Hi,

I’m trying to dock a short polymer (~1.2 kDa) to a protein based on NMR data (chemical shift perturbations). From the documentation it seemed quite straightforward, but I keep running into problems when submitting my polymer structure.

HADDOCK interface: HADDOCK 2.4, easy interface

The protein PDB file:
The protein file is a .pdb I’ve downloaded from PDB. The original file contains two protein molecules defined as chain A and B. I simply removed one of the chains to get a file with only one protein structure. This file seems to be accepted by HADDOCK.

The polymer PDB file:
The polymer comprises 13 identical subunits made of a hydrocarbon backbone with carboxylic acid side chains. Each negative side group has a sodium ion as counter-ion. The structure was made in Avogadro and saved as .pdb. The structure looks fine in PyMOL.
When uploading the structure to HADDOCK I get the following error:

"Error in PDB file. Your PDB contains multiple residues with number 1 in chain or duplicated atom names.

HETATM 2 C UNL 1 -3.110 -42.084 -0.234 1.00 0.00 C (Offending Line) <–
ATOM 32 N ARG 3 11.281 86.699 94.383 0.50 35.88 N (Example Valid Line)"

As I understand it from reading other threads, this is caused by Avogadro not separating the different subunits into “residues”, but listing them as all belonging to the same “residue”. I can probably fix this, but I was wondering, do you have any extensive documentation on what the expected input file should look like and what “flags” are accepted? E.g. is the “UNL” flag accepted?
I’m asking because I’m a bit worried that something else is wrong with my file. I was trying out your pdb-tools web interface after seeing this thread; Error in load pdb file for HADDOCK, but when I preview my submitted file the preview shows the atoms of one subunit, but no bonds or any of the other subunits. Is this because of the above all-residues-are-numbered-the-same problem?

Then I just have another question: Can HADDOCK take surrounding ions, i.e. my Na+ ions, into consideration when performing the docking - or would it be better to remove them?

I’m thankful for any help I can get on the matter. Also, if anyone has any software that they think is better than Avogadro for building small molecules, I would appreciate any and all recommendations. Avogadro seems to randomly crash for me on Mojave. (I used to use Chemdraw, but they seem to have removed their 3D tools for OS X.)

Regards,
Therese

2

Hi! Check the HADDOCK 2.4 Library for a list of supported modified amino acids and ions.

As I understand it from reading other threads, this is caused by Avogadro not separating the different subunits into “residues”, but listing them as all belonging to the same “residue”.

This could indeed by the cause of the Your PDB contains multiple residues with number 1 in chain or duplicated atom names error.

3

Hi Therese

I am afraid this is not going to be possible… or complicated.

Basically your polymer must be defined as one single residue with unique atom names, all as HETATM
The ions can be defined as separate atoms, each with its own residue number, but you will have to define distance restraints to keep those in place.

Polymers is not something we are supporting in principle.

4

Thank you for your replies.

Hmm, I understand. Regarding “… one single residue with unique atom names …”; would it work to define them as C1, C2, C3, etc.?

I could purge the system of the sodium ions. There’s no real way for me to know if they are really part of the interaction anyways.

Otherwise I might be able to get an idea of the interaction by docking a poly-Asp peptide. The chemical properties of the backbone will be completely different though… I have to think about this…

/Therese

5

Hmm, I understand. Regarding “… one single residue with unique atom names …”; would it work to define them as C1, C2, C3, etc.?

Yes

I could purge the system of the sodium ions. There’s no real way for me to know if they are really part of the interaction anyways.

That would simplify things indeed

6

Hi again,

Thank you for your previous help. Since then I’ve been doing some troubleshooting. So, first; renaming the atoms, fixing numbering, etc. seem to solve it. Avogadro overwrites the changes I make every time I open the file though, so I need to find another software to use when working with the file. The polymer is too long for me to want to write the entire file manually.

I extracted a shorter version of the polymer from an existing PDB structure, similar to what was done in this protein-ligand docking guide: http://www.bonvinlab.org/education/HADDOCK-binding-sites/ . I figure I can just use it for now while solving the problems with generating a file of the full-length polymer.

When following the guide I’m running into a problem though; in order to limit the possible interaction sites I’d prefer to define active residues, but this requires me to also define active residues in the ligand. And the ligand doesn’t have any residues per se. Is it possible to define specific (or all) atoms in a ligand as active?

Best,
Therese

7

Hi Therese

The ligand is one single residue. Simple define it as active.