Arking:JCAOligoTutoria24
This tutorial was adapted from the "Basic Tutorial For DeepView" available at http://spdbv.vital-it.ch/TheMolecularLevel/SPVTut/. I've simplified it somewhat to illustrate the basic operation of the program. For the full glory of DeepView, I recommend following the complete tutorial available at the above URL.
Download DeepView
Go to http://spdbv.vital-it.ch/disclaim.html, accept the disclaimer, and then download the file. On a windows machine, it will come down as a zip file. You should unzip the file, and then double click on spdbv.exe to launch the program. (There is no "installer" for DeepView, and you can launch it from anywhere).
DeepView is often referred to by different names including spdbv, Swiss view, and Swiss-PdbViewer...it's all the same program, just different names.
Download a File
There are several ways to get structure files (.pdb files) for viewing in DeepView. DeepView can access pdb files from the Protein Data Bank directly, but I'll show you how to get to things via pubmed. For this tutorial, we want to work with Hen Egg-White Lysozyme in complex with the trisaccharide inhibitor tri-(N-acetylglucosamine). The ID code for the file we want is 1HEW.
Go to www.pubmed.gov, type "Hen Egg-White Lysozyme" into the search box, put the search pulldown menu on "Structure" and hit GO. This will pull up a lot of hits of different structures of Lysozyme. Each entry is a distinct crystal structure. Often the different hits are mutants of the original, different inhibitors bound to the protein, or maybe just another author's structure of the same protein. Unless you know specifically what ID file you want, you usually use keyword searches and browsing through the list to find what you are looking for. For this example, we know exactly what we want: 1HEW.pdb. So, type "1HEW" in the search box and hit GO. You should get this:
Click on the image, and that should take you to a screen that looks like this:
I've circled the link you should now follow to get to the Protein Data Bank where you can download the file. Click it.
I've circled the link to click to download the pdb file. Click it and save the file to your desktop. Now you can open it into DeepView. Double-clicking the desktop icon might do the job. If not, do "File > Open PDB File..." within DeepView to open the file. There might be some error messages that pop up. If you really want it not to give you these error messages, you can turn them off in the DeepView preferences. Just say ok to get rid of them for now. You should have something that looks like this:
If you don't have that rightmost window, the "Control Panel", you should open it up from the Window menu. Alright, now let's play with it!
Selecting and Displaying
You use the Control Panel to select, display, or hide parts of the model. Different parts of the model such as amino-acid residues or hetero groups like the three NAG residues in the lysozyme inhibitor are all called "groups" in DeepView. Selecting groups does not change the display. But most actions that you take on the model affect only the currently selected groups. The Control Panel lists all the groups in the model and allows you to select, hide, and display them. You may find this Panel imposing at first, but when you get the hang of it, you will be able to control the display and determine the targets of commands very quickly and precisely.
Click anywhere on the Control Panel window to make it active. First, simply scroll down to the bottom of the list to see how many amino-acid residues the protein contains, and to see how the hetero groups are named at the bottom of the list (hetero groups are always listed last). The tri-NAG inhibitor is listed as three residues of NAG, numbered 201 to 203.
Now click and drag, starting on the word GLY4 at the top of the window, dragging down to GLY16 and releasing the mouse button. All group names from GLY4 through GLY16 should turn red. Groups printed in red are now selected. The simplest way to select a small number of residues is to click their names and drag to select a range of them.
Now hit <Enter>. This will hide all groups except those selected. You are looking at residues 4 through 16 of the protein. Notice that a checkmark (on Windows, a checkmark looks like the letter v) has appeared in the "show" column next to the selected residues in the control panel, indicating that they are on display. There are checks also in the "side" column, which means that side chains are shown. A side chain is shown only if the rest of the residue is shown, so you only see the side chains of displayed residues in the graphics window.
Now, let's center the view and play with it. There are different mouse modes that controls what happens to the display when you click or drag in the graphics window:
Click the button for "Scale and center visible groups". For now on, I'll just refer to this as "centering the view". Notice that after centering, the model rotates about the center of the groups on display. Centering the view with this button just does the action--it doesn't change the mouse mode. The next three buttons are different. Go ahead and click on one of them and then click and drag on the screen. This is how you manually zoom in and out, spin the image around, and shift it left and right. Play with that a bit and then re-center the view.
Now let's play with the Control Panel some more. Click the check in the side column next to MET12, to hide the side chain. Click the same space again to replace the check and display the MET12 sidechain. Try the same thing in the show, label (text labels), and surface (dotted van der Waals surface) columns, and look at the effects. Now remove the checks from the label and surface columns.
Again select and display residues 4-16 with side chains.
You can also turn on a whole column-worth of settings at once. To do this for just the selected groups, you click the header word at the top of a column in the Control Panel. Let's do this for "ribn". With residues 4-16 still selected, click on the word "ribn". You should get a little alpha helix showing up. Try spinning the image around, re-centering it to get it all on the screen, and playing with the other mouse controls. Since the view is in 3D, but you are viewing it in 2D, it will often help to wiggle the view around with your mouse to help orient you to what is above and behind your view.
You can also turn on or off an entire column-worth of settings for the entire model with shift-clicks. To remove the labels, surface, and ribbon, hold down the shift key and click any checkmark in the label column (not the word label). Repeat for the surface and ribn columns. When you click on a checkmark while holding down shift, your action removes all checkmarks, and DeepView makes the corresponding changes in the display. You can also turn the whole column on by shift-clicking on an unchecked position in the column. So, you can toggle the view to be all-on or all-off by using shift-click for a particular column of settings.
Next, select and show residues 4-16 with their side chains, and add residues 17 through 22 as follows: click in the show column next to LEU17, drag down the column to GLY22, and release the mouse button. Arrows should appear in the show column next to residues 17 through 22, but these residue names should not turn red. If you drag too far, release the mouse button and then remove unwanted checkmarks by clicking them one at a time. Add checkmarks in the side column for residues 17-22 in the same way. Now residues 4 through 16 are selected and shown, while residues 17 through 22 are shown, but not selected.
Again, display residues 4-22 (all with check in the show column, but with only residues 4-16 selected (red). Notice the + and - symbols above the Control Panel column headings. These act as buttons to turn the column function (show, side, label, and so forth) on or off, but only for the selected groups. Click on them and see for yourself.
For easily selecting a range of residues, click on a name to select the first residue, then hold down shift while clicking the last residue in the desired range. DeepView selects your first and last residue, and all those in between. Try it.
You can also select groups of residues that are not consecutive. With residues 4-16 selected and shown, hold down the control key (ctrl) and click TYR20 and LEU25. This adds residues 20 and 25 to the selection (shown in red). Press enter to add them to the display. These residues are disconnected from the others. If you do not see them, Zoom/Center to center the displayed groups. Click the word side at the top of the Control Panel. This adds side chains to the newly selected groups. Note again that clicking headings in the Control Panel affects the selected residues only.
There are also two narrow columns to the left of the group column. The first column is blank when the current model contains only one protein chain. Usually, you'll get some letters. For 1HEW.pdb, you should see "A"s by each group. These refer to the biopolymer chain. If there were multiple subunits in the file, or perhaps a DNA in there, there will be multiple chain labels for each biopolymer present. If you click on one of these chain labels, you select the entire chain.
The second column contains groups of the letters h or s. Groups labeled h comprise alpha helices, while groups labeled s comprise strands of beta sheet. Click on any h. You have selected the entire helix containing the residue you chose. Center the view to zoom in on the helix. Now find a different helix in the Control Panel list. Hold down ctrl and click any h in your chosen helix. You have added the second helix to the display (without selecting it). Center the view again to see what you did.
Select Menu Basics
The DeepView Select menu provides other commands for selecting.
"Select > All" selects all groups in the model
"Select > Secondary Structure > Helices" selects all the alpha-helical regions of the model
"Select > Seconday Structure > Strands" selects all the strands of beta sheets.
Another useful one is "select > visible groups". This one will select all the groups that are currently visible on the screen.
You can also select everything other than those currently selected with the "Select > Inverse Selection" choice.
There are lots of other selection methods available from this menu, and we'll visit some of them later. You can do things such as selecting all the tryptophan residues, or all the hydrophobic residues.
Coloring
Colors can reveal structural, chemical, and comparative features vividly, and can help you to keep your bearings during complex operations.
Select, display, and center the complete model, without side chains. Checkmarks should only appear in the "show" column.
You can color groups manually using the Control Panel, or using various options within the "Color >" menu. Let's start with the color menu. When opening up a file for the first time, it is very useful to color the chains to get a feeling for what different biopolymer chains are present in the model and how they are oriented. You'd do this by "Select > all" (or hitting <ctrl-A>) and then selecting "Color > Chain". For lysozyme this isn't very interesting since there is only one chain, but for multiple chains they'd all turn different colors.
You can also color things by various biochemical properties from the color menu:
"Color > Secondary Structure" colors helical residues red, beta sheet residues (strands) yellow, and all others gray.
"Color > Secondary Structure Succession" colors helices and strands, but with this command, color reflects the order of each structural element in the overall sequence of residues. DeepView colors the first element of secondary structure violet, the last one red, and the ones in between with colors of the visible spectrum that lie between violet (400 nm) and red (700 nm). The result is that it is easy to follow the chain through the protein -- elements of secondary structure are colored from the N-terminal to the C-terminal end in the order violet, blue, green, yellow, orange, red.
You can also color things manually from the Control Panel.
On the Control Panel, notice that, next to the column heading "col", there are the letters BS. This indicates that coloring commands will change backbone and sidechain displays only. Now pull down the Color menu, and notice that the first line brings down a submenu of "act on" commands. This submenu determines which display elements (backbone, sidechain, ribbon, and so forth) will be changed by subsequent Color-menu commands. (On Macintosh, this submenu is also located below the col heading of the Control Panel. Find it by clicking the small black triangle below col.)
Color: act on Sidechains Notice that the letters BS are now replaced by S, indicating that any subsequent Color commands will now change only the colors of side chain, which backbone color is unaffected.
Color: Type This command recolors the side chains all residues according to chemical type. Note that it colored only the sidechains because the Color menu is set to sidechains only; mainchain residues remain colored by secondary structure. Nonpolar sidechains are now gray (other side chains are not shown because you selected and displayed nonpolars only). Look at the control panel to see the colors assigned to other types of residues. What are the colors for basic (positive), acidic (negative), and polar residues? Now look at the model. Notice that most of the gray side chains on display cluster in the heart of the molecule, as you would expect, because most hydrophobic side chains in a water-soluble protein like lysozyme are buried.
Select: Group Property: Acidic <control> Select: Group Property: Basic On the second command, press the control key first, and hold it down while pulling down the menu. Using control with a Select menu command adds groups to the selection, without un-selecting others. (Recall that the control key has a similar function in the Control Panel.)
Linux: Control key function may be missing.
heading: side This adds side chains for selected groups, and removes all other side chains. Look around the model. Most of the red and blue side chains shown are on the surface, as you expect for charged side chains in a water-soluble protein. Note that the nonpolar side chains are no longer on display, because you did not hold down the control key during the Select: Group Property: Acidic command above.
Can you show only the polar sidechains on the backbone? To DeepView, "polar" means polar but not charged.
Again select, display, and center the complete model, with side chains.
Color: act on Backbone + Sidechains This action resets the Color menu so that subsequent Color commands affect backbone and sidechain displays.
Color: Accessibility This operation may take 10 or more seconds, depending on the speed of your computer. (During slow operations, DeepView shows a progress bar to let you know everything is all right.) After the calculation, you see the model in colors ranging from violet to orange. The color of each residue is based on the percentage of its surface area that is exposed (accessible) to the surrounding solvent. The least accessible (buried) residues are colored violet. To residues of higher accessibility, DeepView assigns colors of higher wavelength in the visible spectrum (the color violet is about 400 nm, and red is about 700 nm). So in this display, the colors of the rainbow reflect the exposure of the residue to solvent: violet residues are the least exposed, red residues the most exposed. Find tri-NAG (showing surface is a good way to find a group quickly). Which NAG residue is most accessible?
Also notice that the colors of the visible spectrum are used here in a manner that is common to coloring for many quantitative features: violet for the lowest values (in this case, % exposure to solvent), and red for the highest values, with intermediate values represented by colors between blue and red on the visible spectrum. (In Color: Secondary Structure Succession, the quantitative feature was residue number in the sequence).
Technical Note
To be more precise, the accessibility of each residue is computed as this ratio: (exposed surface area) / (maximum possible exposed surface area)
The maximum for residue X is defined as the exposed surface area of residue X in the pentapeptide gly-gly-X-gly-gly in fully extended conformation. To make this calculation, DeepView adds 1.4 angstroms (approximate radius of a water molecule) to the radius of every atom, computes their surfaces, eliminates all overlapping surfaces, computes the exposed surface area for each residue, computes the ratio of that surface to the maximum possible, and assigns colors on this scale: 75% of maximum exposed: red; 37.5% exposed: green; 0% exposed: violet, with intermediate colors for intermediate values. You can see why even DeepView, which is a very fast program, takes a few seconds to carry out this calculation. (Well, that's what I said a few years ago—but as you can see, the calculation time is now less than one second, and no progress bar appears.)
Select and center TRP62 by option-clicking the residue in the control panel. With only TRP62 selected and the whole model shown, click on the surface header (small cluster of dots, with a v -- for van der Waals -- at its lower right) in the Control Panel. TRP62 is now shown with a dotted surface that represents the van der Waals radius of each atom. Zoom in so that you can see this residue and its surroundings clearly. Notice that it lies just below the surface of the enzyme, almost completely surrounded by other groups.
Linux: There may be no counterpart to the option key on the Linux keymap.
Notice that, just beneath the surface heading, there is a small black triangle. This is the handle for another pull-down menu. Click on the triangle to pull the menu down, and select accessible. Notice that the v becomes an a -- for accessible. Again, with the whole model on display, but only TRP62 selected, click on the surface heading. A small region of surface dots appears, outside the van der Waals surface. This surface is that portion of the surface of TRP62 that is solvent-accessible, or exposed to the surrounding medium or solvent. You can see that only a small fraction of the TRP62 surface is solvent-accessible, in keeping with the violet color assigned when you colored it by accessibility.
The van der Waals surface lies at a distance of the van der Waals radius from the center of the atom directly below its surface, and includes the full surface of all selected atoms, except where their surfaces overlap. In contrast, the solvent-accessible surface lies above each atom at a distance of the van der Waals radius plus 1.4 angstroms, and includes only the surface with which a spherical solvent molecule of radius 1.4 angstroms could come into contact. Any other part of the surface is hidden from the solvent by surrounding residues. Like coloring by accessibility, showing the accessible surface requires a lengthy calculation, and can take a long time for the entire molecular surface of a large protein model.
Turn off both the van der Waals and accessible surfaces. Center the model, still colored by accessibility.
NOTE: If the dots on van der Waals and accessible surfaces are too sparse for your liking, increase their density with Prefs: Display. Changes take effect the next time you move the model. I use 8 for both vdW and surface dot density.
Display: Slab This command reduces the display to a thin slab centered at the middle of the molecule. DeepView shows only the groups that lie within this slab; the program hides groups that lie in front of or behind this slab. With this display of the molecule in cross section, you can see that residues in the heart of the molecule are not accessible to solvent (dark blue), while surface residues are accessible. Rotate the molecule to see that this is true for all the surface.
You can change the thickness of the display slab (to display a thinner or thicker section of the molecule) using the mouse. Hold down shift and left-click drag to the left or right. Moving to the left thins the display slab; moving right thickens it. You can also move the slab toward or away from you, keeping the same thickness of cross section, by holding down the shift key and left-click dragging the mouse forward or backward. Moving the mouse away from you pushes the slab farther away, revealing groups more distant from you; moving the mouse towards you pulls that slab forward, revealing groups in front. This works no matter which of the manipulation buttons -- rotate, translate, or zoom -- are currently selected. Zoom/center to return the slab to the center of the model. Then turn off slabbing by again selecting Display: Slab. Slabs are very handy for eliminating background and foreground groups and exposing the groups you want to study.
How can we color individual residues, or complex selections of residues? The small colored squares in the col column of the Control Panel provide a means to color individual residues, while the col heading at the top allows you to choose a color for all currently selected residues. Display the whole model with side chains, and select all groups except tri-NAG. (There are several ways to make this selection. One quick way is to select NAG 201-203 and then choose Select: Inverse Selection.)
Next you will change the color of the selected groups. Click to activate the Control Panel.
heading: col You will see a standard Macintosh color wheel, and below it, a shade bar. If necessary, slide the pointer on the shade bar to give the wheel bright colors. Then click in the green region of the wheel. Click OK. Now the whole model, except for tri-NAG, which is not selected, is green.
Linux and other versions: Your familiar color picker should appear.
Next, color tri-NAG red, as follows. Select NAG 201-203.
heading: col Set shade and color to a bright red that will show up well against the black background. Click OK. Now the green enzyme sports a red inhibitor.
You can also color any individual residue by clicking on its individual color box, regardless of whether it is part of the current selection. Try this by coloring LEU129, the C-terminal residue, yellow.
Do you see any other yellow bonds in the model? Identify these yellow structures. Center on one of them as follows:
Click the fourth icon from the right on the graphics window (it has an eye and 4 arrows on it). As instructed below the icons, click an atom -- choose one of the atoms connected by a yellow bond (in stereo, remember to click in the left image). The atoms you chose becomes the center of rotation and display. Zoom in to get a closer look -- slabbing might help.
Select: All Color: CPK This action returns all groups to standard or CPK colors: white for carbon, red for oxygen, blue for nitrogen, and yellow for sulfur. This color scheme will help you to identify the yellow bonds, which are the common disulfide cross-links between cysteine residues. Disulfides, which stabilize the three-dimensional structure of proteins, are found mostly in extracellular proteins like lysozyme. Life is rough outside the cell, and disulfides make proteins more rugged.
By the way, the way to color only selected groups CPK is to click col, and then click Cancel on the color wheel dialog.
Test your DeepView skills:
Make a model in which lysozyme is shown only as a ribbon, colored by secondary structure, and tri-NAG is shown in normal (CPK) colors with a dotted van der Waals surface. Your model should look like this (but in a different orientation):
Linux: Ribbons don't seem to be able to take up any kind of colour scheme. The closest approximation to the above picture is one with the wireframe coloured in a similar manner to the ribbons.
