3
$\begingroup$

Problem: I have a PDB file, with a monomer, but I would like to show the entire structure - which is trimeric - but I don't understand how to merge/build or combine the monomeric units to its full structure in COOT, SWISS-PDBviewer or pymol?

Here is a link to the protein complex I am looking at

I would appreciate: Guidelines on how to do this (preferably in coot, pymol or Swiss-PDBviewer), or referral to a tutorial that actually goes through this would be great!

For reference: I found some descriptions on how to "build/merge" with Swiss PDB viewer here. And in threads from this discussion I found descriptions of how to do it with other molecular graphic tools as well (although I don't completely understand it)

What I have tried: In the tutorial for the SWISS-PDBviewer ( link ), I am able to follow some of the instructions, but I can't follow them for long:

  • I upload three of the same PDB file in SWISS PDBviwerer (they are layered on top of each other)
  • I can access/see the layers in the "layers info"
  • I am told to open the "text icon" to see the PDB text file, and look for "mtrix" rows - which should come right before the "atom" rows. As cited from the instructions:

" Scroll down the pdb file until you find MTRIX lines (they are just before the ATOM lines). You can see 9 lines MTRIX. They represent three transformation matrices, and allow you to build the non-crystallographic symmetries of the protein"

I cannot find the "mtrix" rows in the PDB-text file, and am very unsure of how to follow the next instructions (link) after this as well. I am not able to click anywhere in the text file, then I get the error: "sorry can't recognize any clickable information under this pointer..

This is what I see above the "atom" row:

enter image description here

$\endgroup$
  • $\begingroup$ I also found that the trimeric state can be downloaded as "biological assembly", however, I would still like to know how to do this on my own. $\endgroup$ – CuriousTree Dec 9 '15 at 13:41
1
$\begingroup$

The PDB-file only contains the asymmetric unit, and no information on a potential biological relevant multimer. So you will need to get information about the state in solution experimentally.

That being said, one can often deduce from the size of the contact surface whether you have a potential quarternary structure. You can do this by hand, but the PISA-server is quite helpful there, and it will also output the multimer PDB-file:

http://www.ebi.ac.uk/pdbe/pisa/

If you just want to apply symmetry operators to your asymmetric unit in order to create symmetry mates, you can do so in

Pymol:
- click Action/generate/symmetry mates/within X Å
- from the command line:
        symexp prefix, selection, cutoff
        , e.g. symexp sym,1GVF,(1GVF),5
   see also http://pymolwiki.org/index.php/Symexp

Coot:
- if you just want to look at the symmetry mates, activate View/Cell&Symmetry with an appropriate radius and/or select "Symmetry by molecule"
- if you would like to actually create the symmetry mates, try Extensions/Modelling/New Molecule from Symmetry Op  - however you will have to specify the SymOp manually.

You will need to have the CRYST1-record in your PDB-file with the correct space group.

$\endgroup$
  • $\begingroup$ Remark 350 can be used to describe the biological unit. $\endgroup$ – Praxeolitic Mar 20 '18 at 19:10
2
$\begingroup$

There are no MTRIX cards for that molecule. This seems like the easiest path:

http://www.rcsb.org/pdb/explore.do?structureId=4G3Y

Download Files -> Biological Assembly 1

i.e.:

http://www.rcsb.org/pdb/files/4G3Y.pdb1.gz

And read that file into Coot (should you wish)

$\endgroup$
  • $\begingroup$ Thanks for the link, but as i commented on my own post - I do understand that I can get the biological assembly (whole structure). But what is an mtrix card? and why is it not possible to combine the monomers (due to lack of mtrix) ? I am more interested in how I would do this on my own, regardless of that the whole PDB structure exists. Are you saying it is not possible with this file? $\endgroup$ – CuriousTree Dec 9 '15 at 18:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.