After this practical you will:
In this tutorial you you will use the Swiss-PdbViewer (also known as DeepView). If you need help using this program during the tutorial, please ask. If you need help when working on your own, a useful starting point is the Swiss-PdbViewer User Guide.
This tutorial contains some step-by-step instructions for performing some basic modelling operations using the Swiss-PdbViewer. The exercises in this section are very artificial, but should help you to become familiar with the program.
An assignment will contain a more realistic scenario: you will take a protein sequence and create a model three-dimensional structure for that sequence based on the experimentally determined structure of a protein with a similar sequence.
To run the Swiss-PdbViewer, type:
spdbv
The are based on the Swiss-PdbViewer Tutorial on Building Loops and the User Guide section on Mutating Amino-Acids.
We shall remodel the loop between crambin's two helices using the "Scan Loop Database" method. Then we shall mutate some side chains. Crambin is chosen for this exercise because it is a small protein, and that makes it easier for us to focus on the modelling steps and the regions being remodelled.
Locate the loop to be remodelled.
From the "File" menu select "Open PDB File..." and open the Protein Data Bank file containing the structure of crambin (PDB Id: 1CRN).
So that the main chain can be seen more clearly, remove side chains from the display. This can be done by placing the cursor anywhere in the column under the heading "side" in the Control Panel window, and then clicking the right button on the mouse. Note that individual side chains can be toggled on and off by clicking the left mouse button in this column. The side chains for a range of residues can be toggled on and off by pressing the left mouse button in this column and dragging the cursor up or down the column.
Crambin contains two helices and two strands. Rotate the molecule and observe these secondary structure elements. To have these regions highlighted, use the "Color" menu and select "by Secondary Structure".
For this exercise it will be convenient to use different colours for the two helices, so select "by Secondary Structure Succession" from the "Color" menu. The first helix is now light green, and the second is yellow.
Select anchor residues.
We shall use residues Leu18 and Ala24 as our anchor points. Identify these residues by placing labels of their CA atoms. This is done easily by placing the cursor in the appropriate rows of the column under the heading "labl" in the Control Panel window and then clicking the left button on the mouse. Rotate the molecule and observe the positions of these residues with respect to the loop.
Search fragment database for candidate loop conformations.
Select "Scan Loop Database..." from the "Build" menu. You will now be prompted to select the two anchor residues. Select the first anchor point by using the left mouse button to click on one of the atoms in Leu18 in the main graphics window. When prompted to select the second anchor, click on one of the atoms in Ala24.
The program will now search a database for candidate loops that can be used to remodel the main chain region between the two chosen anchor points. After a short delay, a window will appear with one row of figures for each of the candidate loops. The loop corresponding to the row of figures shown in red is currently displayed in the main graphics window. You can view other candidate loops either by clicking on a row of figures using the left button on the mouse, or by using the up-arrow and down-arrow keys to move through the list of fragments.
Select a loop conformation.
The loop that we are trying to remodel has the sequence "PGTPE". Step through part of the list of candidate loops and compare the sequences of the fragments with the sequence that we are trying to remodel. You might find a fragment with an identical sequence. From which structure has that fragment been taken?
Clicking on "FF" or "PP" above the line in the results window will cause the list of candidate loops to be sorted according to a different criterion. Sort the candidate loops by "FF". Notice that the fragment with the target sequence does not have the lowest "FF" value. Select the loop with the lowest "FF" value (this is not a general recommendation; indeed in this case the fragment that has an identical sequence would probably be a "better" choice).
Evaluate the model
We shall now compare the loop that you have chosen with the actual loop in the original crambin structure. To do this, select "Open PDB File..." from the file menu, and choose "1CRN". We now have two "copies" of crambin in the viewer. From the "Window" menu select "Layers Infos". We can toggle visibility of the two crambin structures by clicking the left button on the mouse when the cursor is in the column with the heading "vis" in the Layers Infos window.
By clicking on the molecule identifier ("1CRN") in the "Layer" column we can choose which of the two structures will be the active one that shown in the Control Panel. Select the second copy of crambin and remove its side chains from the display. Rotate the structures and locate the loop between the two helices. Observe any differences between the candidate loop that you selected, and the actual main chain conformation from the Protein Data Bank file.
Select the first copy of the crambin structure (the one with the remodelled loop) in the "Layers Infos" window, making it the active molecule again. Use the up-arrow and down-arrow keys to view some alternative candidate loop conformations and compare these visually with the actual conformation.
If you want to save the results of modelling then you can use the "File" menu and select "Save -> Layer...". You will be prompted to give a name for the PDB-format file to which the contents of the selected layer will be written out. There is no need to save the remodelled crambin structures.
We shall now mutate some of crambin's residues. Use the Layers Infos window to make only the third "copy" of the crambin structure visible, and make this the active structure, then repeat the following steps for four or five residues.
Mutate an amino-acid residue.
Click on the "MUTATE" tool at the top of the display window (this tool, and the others, are shown in the Tools section of the Swiss-PdbViewer User Guide).
Select the residue be mutated (e.g. Glu23) by clicking on an atom in the main display window.
A menu of the 20 different residue types will be displayed; choose the new type for the residue (e.g. "TYR").
The program will then fit all rotamers of the new type onto the main chain of the chosen residue, and will compute a score for each. The rotamer giving the lowest score will be displayed.
On the status line you will see symbols resembling "<" and ">". Clicking on these will allow you to step through the alternative rotamer conformations. Any bad contacts involving the remodelled side chain will be shown as dashed lines in the graphical display.
If you click on the "MUTATE" tool again, a pop-up window will give you an opportunity to accept or discard this mutation.
Adjust side chain torsion angles.
Click on the "TORSIONS" tool at the top of the display window.
Select the side chain to be adjusted (in this case, adjust the mutated side chain) by clicking on an atom in the main display window. Several symbols resembling "<" and ">" will be displayed. Clicking on these will allow you to rotate around one of the side chain's bonds. The top pair of symbols correspond to the chi-1 torsion, and the lower ones correspond to torsions that are successively further from the main chain. As the torions are changed, the bad contacts will be updated. Try to eliminate all bad contacts.
If you click on the "TORSIONS" tool again, a pop-up window will give you an opportunity to accept or discard these torsion adjustments.