• Open 3 separate tabs, or windows, all viewing the PDBsum website at http://www.ebi.ac.uk/thornton-srv/databases/pdbsum/
• Enter one of these PDB codes into the PDB code box in each separate tab or window and click Find:
  • 1sqc - (mainly alpha class, a bacterial isomerase)
  • 2bbk - (mainly beta class, a bacterial methylamine dehydrogenase)
  • 1wvk - (few secondary structures class, arabidopsis, unknown function)

On the next page, you will see a small brightly coloured Ramachandran plot on the right hand side under the title Procheck. When you hover the mouse over this you will see a larger version.

• Click the small Ramachandran plot image to change to a page with the full-size Ramachandran plot.

In this colourful Ramachandran Plot, the most favoured regions are shown in red and brown, while less favourable regions are shown in off-yellow and yellow. Disallowed regions are shown as the pale cream colour.

Superimposed on this background you will see squares and triangles representing the phi/psi combinations of the amino acids in the specific protein of interest. Triangles represent glycines while the other amino acids are shown as squares. Most of the amino acids are shown in blue; those that are in red are in less-favourable regions of the plot.

In the top-left-hand corner of the page, you will see a cartoon image of the protein so you can compare the structure with the Ramachandran plot.

Preliminaries

• Go to PDBsum at http://www.ebi.ac.uk/thornton-srv/databases/pdbsum/
• Enter 1DVJ as the PDB ID to search for information on the decarboxylase.

Check out the Ramachandran plot - what, if anything, can you deduce from this about the 3D structure of the protein, or its quality?

Visualisation with the PyMol program

• Start the PyMol software by clicking on the Windows menu (bottom left of the screen), selecting 'All Software', opening up the section containing programs starting with the letter 'P' and selecting PyMol.
  Note that there are several versions of PyMol available. You probably want the Default 3D Quad buffer one, but it depends on the computer you are using. You do not want the Viewer Only version.

When you start PyMol, a window will open with the PyMol Command region at the top and the Visualisation section at the bottom.

• From the PyMol File menu, select Open and in the resulting dialogue, select pdb1dvj.ent. If this file does not appear in the list of files, then you will need to download it and save it to your desktop; you can then read it into PyMol from there.

You are now viewing the structure of orotidine 5'-monophosphate decarboxylase using a 'ribbon' model in which beta-strands are shown as arrows, alpha-helices as coils and other residues as lines.

You will now change the view to a 'wireframe' model in which each bond is indicated by a thin line.

• On the right of the structure display, you will see the word 'all'. To the right of that, click the letter 'S', then hover over 'as', slide the mouse to the new popup menu and select 'wire'.

We now wish to display the non-protein atoms as spacefilled spheres.

• Click in the command line entry region (next to the 'PyMol>' prompt) and type:

  show sphere, hetatm

This changes the representation of the 'hetero' atoms (i.e. non-protein atoms) to spheres. If this didn't work, make sure you typed 'hetatm' and not 'hetatom'!

You should now be able to see the ligand (spacefilled) against the protein (wireframe). However, many red spheres will also appear. These are water molecules which have crystallized with the protein.

• Remove the waters from the display as follows:
• On the right of the structure display, you will see the word 'all'. To the right of that, click the letter 'H', then select 'waters' from the popup menu.

To zoom in on the structure, press the right mouse button while you move the mouse up and down.

We are now going to return to the cartoon view of the protein and colour it by secondary structure

• On the right of the structure display, you will see the word 'all'. To the right of that, click the letter 'S', then hover over 'as', slide the mouse to the new popup menu and select 'cartoon'. Note that the spacefilled display of the non-protein atoms will disappear when you do this - you can repeat the two steps above to restore them.
• On the right of the structure display, you will see the word 'all'. To the right of that, click the letter 'C', then hover over 'by SS', slide the mouse to the new popup menu and select the first of the three colouring schemes where Helix is in red.

You will now have the protein rendered using cartoons to show the secondary structure elements. The alpha helices will be shown in red, beta strands in yellow and other parts in green.

The crystal structure contains 4 copies of the protein. We will now select just one of those to view and switch the other ones off.

• In the Command Line window, enter the command:
  select notA, chain B:D
  This creates a selection containing chains B-D and calls it 'notA'.
• On the right of the structure display, you will see the word '(notA)'. To the right of that, click the letter 'H', and click 'everything'
• Click on the black background of the display to remove the pink squares.

Now we need to centre the rotation about the remaining chain A:

• In the Command Line window, enter the command:
  select chainA, chain A
  (Note there is no space in the first 'chainA', but there is a space in the second 'chain A'). This creates a selection containing chain A and calls it 'chainA'.
• Now enter the command:
  origin chainA
• Click on the black background of the display to remove the pink squares.

You will now find that when you move the mouse pressing the left mouse button the centre of rotation will be the centre of the remaining chain A. You can move it to the centre of the display by pressing and holding the middle mouse button.

Try a different representation of the protein structure - a C-alpha backbone plot:

• On the right of the structure display, you will see the word '(chainA)'. To the right of that, click the letter 'S', then hover over 'as', slide the mouse to the new popup menu and select 'ribbon'.

To restore the display of the ligand associated with Chain A, we can create a selection: Note that this may not work as described depending on the PyMol version! There is an alternative shown.

• At the command prompt type:
  select hetA, hetatm and chain A and not resn hoh
  This selection, which we have called 'hetA' contains non-protein atoms that are part of chain A and which do not have the residue name 'hoh' (i.e. water)
• If this didn't work for you (PyMol responded with Selector: selection "hetA" defined with 0 atoms), then, at the command prompt type:
  select hetA, hetatm and not resn hoh
  This selection, which we have called 'hetA' contains non-protein atoms that are part of any chain and which do not have the residue name 'hoh' (i.e. water). This will result in all ligands being displayed instead of only the chain A ligand, but it doesn't matter.
• On the right of the structure display, you will see the word '(hetA)'. To the right of that, click the letter 'S', then hover over 'as', slide the mouse to the new popup menu and select 'spheres'.

You can access LIGPLOT diagrams from the PDBSUM web pages that you used earlier.

• At the top of the 1DVJ entry page in PDBSUM, click on the Ligands tab.

You will see a LIGPLOT schematic of the inhibitor-enzyme interactions.

• Go to the CATH database at http://www.cathdb.info/
• Search for the PDB code 1dvj.

On the next page, you will see a 'domain ID' for your protein (1dvjA00) and the corresponding CATH code 3.20.20.70

• Click on the domain ID (1dvjA00)

This will retrieve a classification hierarchy for orotidine 5'-monophosphate decarboxylase. You can see that its topology (3.20.20, the overall shape and way in which secondary elements are connected, i.e. the protein fold) is classed as a TIM barrel. This is a so-called ‘super-fold’ – a highly populated protein fold which contains many proteins which do not have any homologous relationship (i.e. the fold has evolved several times independently)

• Click on the CATH code 3.20.20 representing the topology level of the classification to see a list of all the homologous families which have the TIM barrel fold.

The blue circle represents the 3.20.20 TIM barrel fold. The orange ring around it broken into segments represents each of the homologous families that adopt this fold.

• Hover over the sections of the orange ring to see the different families.