Preamble: This tutorial was written by Dr. David Hackney; it has
been used by students in biochemistry courses at Carnegie Mellon University for
several years. It was adapted for this HTML version by Dr. W. McClure. If
your monitor size permits, you can work through this tutorial by scrolling in the Netscape
window in half of the screen, while you operate in the two 2RasMol
windows on the other half of the screen. Alternatively, print these pages and use
them with only the RasMol windows active.
Page Appearance: The tutorial uses two different fonts and both are under your
control. (Netscape 3.01: look in the General Preferences... menu under Options.
Netscape 4.05: look in the Edit menu under Preferences..., then Appearance
and Fonts.) In addition to this "Proportional Font", which is used for
explanations, all of the RasMol commands and command arguments are in "Fixed
Font", e.g.selectatomno=2. The Netscape
default is 12pt Courier, but any other font and size combination can be
selected to distinguish the explanations from the commands.
The first use of each command in the text is linked to a specific section of the on-line RasMol v2.6 Manual, where additional information on
the commands and other features are explained. (Referring to the Manual is unnecessary the
first time through the tutorial. When you are familiar with the basics, redo the tutorial
and consult the Manual to learn more about the variety of each command.)
PDB & Script Files: The two script files and the coordinate file used in this
tutorial can be copied from the RasMol Tutorial Files
page. That page also has a link to the RasMol Homepage where the RasMol program and
supporting documentation can be downloaded.
CMU students and faculty can copy these files and the RasMol program from several course
folders on the AppleShare BioServer. Contact W.
McClure or D. Hackney for details.
RasMol is a molecular visualization program written by Roger Sayle. We will
use it for studying protein and DNA structures. RasMol was originally written for Unix
machines, but versions for MS-windows (RasWin) and Macs (RasMac) are now available. The
Mac version runs on any Mac, but the speed is poor on older Macs and I strongly urge you
to use a PowerMac. This tutorial assumes basic operating knowledge of the Macintosh
interface. The program runs faster from the hard drive, so make a new folder on the
desktop or HD and copy to it all of the files from the floppy or from the file server
folder using AppleShare. The program also sometimes gets confused if everything is not in
one folder, so keep the program file and all coordinate and script files in the same
folder.
Start the program, RasMac v2.6. When the program starts up, it opens
two windows. One is for display (the one labelled 'RasMol Version 2.6') and the other is
to enter commands (named 'RasMol Command Line'). You do not need to switch back and forth,
because both are active at the same time. You want to arrange your screen so that most of
it is taken up by the display window, but with the bottom two lines of the command window
visible so that you can see the commands that you type and the responses from the program.
Note that the response time is longer for larger display window sizes because of the
greater number of pixels that need to be calculated.
The Manual has a General Operation
section, including a page on the Pull-Down Menus.
Look at these topics for more details after you have completed the tutorial.
Once you have the program running, click on File in the Menu bar; drag
to Open; and select the file 'ProG-alpha.ent' as the first file to open. This file
contains non-hydrogen atoms for residues 23-37 of protein G, Gronenborn, et al.
(1991). This segment forms an a-helix and is a good simple
structure to start with. You should now see the structure displayed in the display window.
Rotation: Rotations can be done in three ways. One is just to use the scroll bars.
A good way to do a continuous rotation is by clicking on a scroll arrow and holding the
click while the structure spins. Try it. The other is just to click in the display window
and drag the mouse before letting go. Try it with short drags first to get a feel of how
it moves. Dragging the mouse rotates around the x and y axes in the plane. To rotate
around z, hold the Shift and Command (
) keys and move the mouse without clicking (Windows: Shift and
Right click). The third method is in the Command Line with rotatez 30 for
rotation of 30 degrees around the z axis, etc.
Numbers can be positive or negative. Note that if you want to modify a previous
command without retyping the whole thing, you can type one or more 'uparrow's and the
previous command(s) will be available for editing and re-execution.
Translation: Translation in x and y can be done with the mouse while holding the
Command () key.
No need to click the mouse while translating (Windows: Right click). From the Command
Line, type translate y 50,
etc. Remember that you can type reset
if you loose the molecule.
Zoom: Zoom can be done with the mouse while holding the Shift key and dragging
vertically. From the Command Line, it is of the form zoom300 where the
default is 100. Zoom 300 would blow
the image up three-fold3 and zoom 100
would return it to the default. Numbers less than 100 shrink
the image below the default size. The maximum zoom seems to vary with the particular
structure, but if you try to zoom too much, the worst that will happen is an error message
in the Command Line window.
Center: Center fixes the center around which rotation occurs. It has the form center, followed by an expression
to define an atom or a group of atoms (in which case it uses the center of gravity of the
group). Type center 23 to center on the alanine 23
at one end of the a-helix; rotation
should now be centered around that group. You can also type set picking center.
After this, any atom that you click on becomes the center of rotation. To disable, type set
picking without an argument. Type centerwithout an
argument to reset centering to the whole molecule. You may now need to use translate
or reset to reposition the molecule in the middle of the window.
Slab: Type slab50and it will remove the half of the molecule that is closest to the viewer. This
is useful to see inside a large structure. The allowed range is 100 for
view all, to 0 for view none. If Slab Mode is selected (Options menu),
the slab value can be varied continuously by holding the Control key down while
click-dragging the mouse vertically.
Display: By default, structures will first be displayed in the wireframe mode. Click on Display
in the menu and try the other types, e.g. 'Backbone', 'Sticks',
etc. These changes can also be done (with additional options) from the Command
Line. Pick wireframe under Display menu to show a wireframe view and then type spacefill in the command window
and hit return. Now typespacefill off to go back to wireframe.
Note that 'Sticks' is really a wireframe with a thick width.
From the Command Line you can change the width to suit your taste - try typing wireframe
90, return; and then wireframe 50, return and see
the difference. Note that in the Command Line, if you are in spacefill and type wireframe
50, you will not see wireframes, since they will be drawn inside the
spacefill and will not be visible. You need first to turn off spacefill by typing spacefill
off. The default for spacefill is to use the van der Waals values. If you
give a numerical argument, it will fill to a sphere of that size in units
of 0.004 Å. Thus spacefill 100 gives small
spheres and is approximately equivalent to what you get with 'Ball and Stick' from the Display
menu. Note that the backbone90
command draws a rope exactly through the a-carbons
while trace is similar, but is
smoothed.
Color: You can also try some of the other color patterns under Colours in
the Menu. More details on the color
command and Colors and Color Schemes are in
the Manual. The program recognizes both 'color' and colour' as valid spellings. The CPK
scheme is the default pattern in which carbon is gray; oxygen is red; nitrogen is
blue; sulfur is yellow; and phosphorus is orange. This basic color scheme is also often
used to designate charge since red oxygens are found at negative carboxylate groups and
blue nitrogens at positive amines.
The Mac version has a washed-out blue for nitrogens that I find hard to see. Try using a
customized color scheme that I wrote as a script (a file containing a list of commands
that can be executed). To evoke it, type script
usa in the command window and hit the return (the 'usa' is the name of the script
file that calls up a true red, white and blue scheme). You can also change the background
color from black. Try typing background
white and return. It is best to use a white background like this if you are going to
print to a black & white printer. For different effects, try background
cyan or magenta, etc. A number of other color
schemes are available under the Menu item, Colours. See the Manual for definitions
of the schemes. Try the command color charge for one example
with hydrophobic groups colored red, or color structure to color
code a-helix, b-sheet, etc.
You can return to normal by using command, script usa or color
cpk. The script 'usa-selected' is similar to 'usa', but
only changes the color of the currently selected atoms (see below).
Picking and Labeling: To find out what is what (often confusing in a big molecule),
you can click on any atom and a message will display in the command window telling you
what atom it is. This includes the unique atom number assigned to it in the PDB
file, and general information such as the amino acid type and residue number (e.g.
'CA' is the C-alpha of Ala 23 and atom
number 2, etc.). Try picking atoms to identify Ala 23
at the N-terminal of this peptide and Asn 37 at
the C-terminal. (The C-Alpha is atom number 111.) Note
that individual atoms are not displayed in 'Ribbon' or 'Backbone'
display and so picking does not work. If you type set pickinglabel,
clicking on an atom now writes the label on the structure and not in the Command Line
window. Reclicking a labeled atom turns the label off. To stop further
labeling, typeset picking without an argument. Typing labeloff removes all
existing labels. Typing labelon, labels all
selected atoms, which is usually too confusing to be of use unless only a small number of
atoms is selected. The size (e.g. set
fontsize10) and color (e.g.color label
green) of the labels can be varied. (See the Manual for details.)
Distances and Angles: To activate distance measurements, type set
picking distance on the Command Line. Now clicking on two successive atoms
also reports the distance between them. Typing set picking angle
gives the angle between three successive atom picks. Typing set pickingtorsion gives torsional angles of a four atom set. To stop, type set
picking ident to return to default of atom identity or set
picking off to totally disable.
Select/restrict: It is often necessary to isolate part of a large structure to
simplify the view. This is done by selecting a subset of atoms for subsequent operations.
For example, you may want a ribbon to mark the backbone structure and then have only the
side chains of particular interest displayed as sticks coming off the ribbon. Another use
is to have the protein or parts of it in stick mode, while a bound ligand is spacefill.
More simply, you may want the whole molecule to be colored CPK and then
to turn the lysines green so you can find them easily. To try this, type select lys
and then type color green and lysines should turn green. The
commands select and restrict are similar, butrestrict erases from display everything that you do not specify, while select
leaves the display unchanged. Type restrict lys and only the
lysines should be left displayed. Now typeselect all. Note that
the display has not changed sinceselect only selects and does
not produce action (select also did not do anything to the
lysines above until you told them to turn green). Type wireframe 50
and it will display the selected 'all' as wireframe. Note that most commands only work on
what is currently selected! If you give a command and nothing happens, one common cause is
that you only have a limited number of atoms selected. Type select alland
retry the command.
Expressions: Atom expressions
are used to specify groups of atoms for select and other commands to operate on. There are
a number of Predefined Sets such as none,
all, selected, protein, dna, backbone, sidechain, nitrogen, water, hetero (anything that
is not protein or dna such as ligands and metals, etc.) as well as all of the amino
acids and bases (lys, ala, etc.). Boolean operations are allowed.
This is best shown by examples of expressions - followed by an explanation of what each
specifies:
lys and sidechain - the sidechains of all lysines (would not
include the backbone).
not hetero - everything else, e.g.protein
and DNA.
24, 30-32 - residues 24, and 30, 31
and 32. Note that this will select these residues for all chains
including multiple subunits and both DNA and protein.
24 or 30-32 - same as above.
24 and 30-32- selects nothing since nothing is
common to residues 24 and 30-32.
45 and protein - residue 45 of
the protein only. Use if both protein and DNA are in the
structure.
45 and dna - nucleotide 45 of DNA
only. Use if both protein and DNA are in the
structure.
none- the null set.
all- all atoms.
resno<30 - residues 1-29.
atomno=44 - atom number 44.
elemno>8 - all elements with atomic numbers
greater than >oxygen at 8.
selected or water - whatever was currently selected, plus
water.
within (6.0, atp) - everything within 6 Å of
the ligand ATP.
105:D- residue 105 of chain D
of a multichain structure.
*:B - all atoms in chain B of a multichain
structure, or more tersely, just :B.
hetero and not water - the ligands without the waters,
if present
Note that * is a wild card for a whole field and "?" is wild for a
single character.
To avoid having to retype a long string repeatedly, you can give a selected
group a name and then refer to it by name. If you type define neg (glu or asp),
then this will define the name 'neg' to mean all glu and asp residues. You can later do select
neg and then color green to color only glu
and asp residues green.
Note that backbone is a predefined set, which should not be
confused with the 'Backbone' option under Display in the Menu. The
Display menu Backbone produces a rope like object that runs
thought the backbone positions. Try typing select all, and then wireframe
10 to get a standard view and then typebackbone 80
to turn on a fat backbone rope. Then type backbone off to turn it off. Now type select
backbone and then wireframe 70 and see the
difference. Also try turning on and off strands and ribbon
from the Command Line.
H-bonds: To show H-bonds, type hbonds in the command window. This
works best if you are in sticks orwireframe 50,
etc., so that the real bonds are thick and the H-bonds are thin.
To remove them, typehbonds off. You can control the width of H-bonds
by hbonds 50, etc.
Save: You can save an
image of the current display in any of a large number of formats. Click on Export in the
menu to see the choices. These produce static images that you can view later, but not
change. You can also use write
in the command window to generate a file such as a new pdb file with write pdbfilename. A particularly useful one iswrite script
filename, which writes a script file that recreates the current view. Thus,
if you had a view that you liked and wanted to retrieve later, so as to be able to rotate
it, etc., then you could type write script goodview1 and
any time later, you could type script goodview1 and get your
view back. This also works even if you have closed a file. Just reopen the original pdb
file and type script goodview1. Try it.
Script files: These are simple text files that can be generated or modified with
any word processor. Just be sure to save the script file in a 'text only'
file type and not in the default format for most word processors. Open up the script 'usa'
with a word processor and see how it works as a model for writing your own scripts. Open
up your script 'goodview1' above to see what it looks like. While you
have a word processor running, also open up 'leu-zipper.ent' to see what
a large pdb file looks like with coordinates for each atom in a defined format. Change to
a non-proportional font such as Courier and widen the page so that the file is aligned
with no artificial line breaks. The 'ProG-alpha.ent' was derived from the
original pdb file by deleting coordinates for everything except non-hydrogen atoms of
residues 23-37. Open it up and see.
One warning, script files that are generated by the write command, reopen the
data file and contain information about path. If you change folders/disks, RasMac will get
confused and not be able to locate your pdb file when you try to run the
script. Either keep everything in one folder or open the script file with a word processor
and change/delete the load and path information. Do not even place pdb
files in a subfolder of the one containing RasMac2.6.
More Now that you know most of the basics, try playing with a more complex file.
(The two coordinate files required in this section can also be downloaded from the RasMol Tutorial Files page.)
1. Use RasMol to open and view 'ProG-all.ent' for the complete protein-G
structure (without hydrogens). Note that only one file can be open at a time; so you must
Close the current view before you can Open a new one (under the File menu). The
current file can also be removed from the Command Line with the zap command. The a-helix
that you have been using actually lies over a b-sheet.
Type cartoon
and then use the following commands:
script usa
select backbone
color structure
This will color code the secondary structure, but leave
the sidechains CPK. Rotate it around to get a feel for the secondary structure. Align it
so that you are looking at the a-helix end-on. Now use these
commands:
define grease (phe,trp,leu,ile,val) and sidechain
select grease
color green
spacefill
This shows that most of them are in the hydrophobic core
between the a-helix and the b-sheet
and that they are away from water. Next do:
select all
spacefill
to get a feel for what is exposed on the surface.
2. You should also look at 'leu-zipper.ent' for a complex of a leucine
zipper transcription factor (GCN4) with a short piece of DNA. This file contains 4
segments for the two protein subunits (C & D) and the two DNA strands (A & B).
Type cartoon and then use the following
commands:
select (leu and sidechain)
color green
spacefill
This highlights the paired leucines in the zipper part.
To change the leucines of protein subunit C only, to red, you can do:
select selected and :C
color red
Now do:
restrict dna
to get a view of only the DNA double helix. You can then
do:
wireframe 70
and then use translate, center, and zoom to get a good
view.
Next try:
cartoon off
restrict (11:A or 31:B)
to isolate a single basepair and then:
hbonds 40
Identify these two bases from their structure and check
by clicking on atoms to see their identity. If you next type:
select dna
cartoon
this single base pair will be placed in context.
3. Now get the coordinates of your favorite protein or nucleic acid, and
view it "your way" in RasMol.
References and Links to other RasMol tutorials on the WWW
1. Gronenborn, A. M., et al. (1991) "A novel, highly stable fold of
the immunoglobulin binding domain of streptococal protein G". Science253, 657-661. PDB file: 2bg1.
2. Ellenberger, T. E., et al. (1993) "The GCN4 basic region
lrucine zipper binds DNA as a dimer of uninterupted alpha helices: crystal structure of
the protein-DNA complex". Cell71, 1223-1237. PDB file: 1ysa.
3. Additional structures can be downloaded from the Brookhaven data base.
Best access is via the WWW using the 3DB
Browser at http://www.pdb.bnl.gov or the PDB Lite server at http://www.pdb.bnl.gov/pdb-bin/pdblite.
4. Dr. Gale Rhodes' RasMol Tutorial for Beginners
at the University of Southern Maine.
Dr. Eric Martz' RasMol
Quick Start at the University of Massachusetts. Includes links to other resources
at his RasMol Home Page.
Back to Molecular Models for Biochemistry at CMU