Modelling Series
Learning about advanced data entry techniques. Exposure to recognizing and correcting data input problems. Modification of data files for specialized needs. Modification of MacroModel minimization output files to take into account solvent effects.
Author:
Susan Jean Johns
Things change and evolve in response to their environment. The same holds true for computer software and databases. The collection of structural data has been going on for over 35 years. In that time the methods for collecting the data has changed and the extent of the user base for the data has grown, as well as the complexity of the structures that can now be determined.
When protein structure identification began, only a limited number of specialists were interested in the results. As more and more molecules were determined, non-crystallographers became interested in the data as it related to their own fields of interest, i.e., secondary structure prediction, folding patterns and binding site interactions. Today, interest in these structures has expanded to include the understanding of functions of other proteins which show a degree of similarity to those of determined ones.
One of the driving forces in this evolution has been the improvement in the technology used to collect structural data. The solution of a protein structure takes only a few years, rather than over a decade, once the crystals can be grown to suitable size and quality. New instrumentation allows greater data resolution for larger molecules than was possible before. NMR is now being used to collect some structural data.
Changes in how structural data is collected causes changes in how a database handles the produced data and distributes it. Once a database settles on a new mode of data reporting, software that uses this database's information then needs updated to handle these changes. Computer software is always behind in adjusting to changes that are created elsewhere and applied without advanced warning.
Protein structures are now being determined and entered into the PDB database at a rate of over 500 new structures a year. This is in comparison to just 10 or so a few years ago. Some of this increase is due to mutation studies on the same protein, however a large portion of increase is simply due to determining more new protein structures. There still is a lag between depositing a structure into the database and its general distribution, although even that appears to be getting shorter.
The structures are now more complex than the ones previously reported, containing more substrates and complexes within the determined result. This means that there are more atoms in a structure than there were before. The size of the ascii version of an x-ray collected data file can now be over 2,000 blocks and contain over 10,000 atoms.
With newer equipment and software to do the determinations, additional changes are taking place. Originally, hydrogen atoms were not included in a data set. The equipment of the time could not determine their positions. Now, NMR techniques can and so they may be included in some NMR determined data sets. This can create problems for older visualization software that doesn't recognize hydrogens as being valid standard atoms from a PDB data set, and thus displays them strangely.
Including NMR-produced data within PDB has caused changes in the way data is handled. New subject areas have been added to show that the data was collected in this manner. The subject area EXPDTA has been created to show that the data has been collected via NMR technology. NMR data can result in a number of solutions to a structure, which are now presented. This can result in extremely long data files. The subject area MODEL has been created to denote one of these possible solutions.
MacroModel is a powerful modelling package with numerous features. It also has limitations. In order to understand the effective usefulness of this program, you must know both its strengths and weaknesses.
Feature: Atom Selection
Characteristic:
MacroModel does not use all of the elements from the Periodic
Table. It has the common elements needed to build carbohydrates, nucleotides,
most organic molecules, and peptides. The software was not designed as an
inorganic modelling tool. It does have a generic symbol for atoms beyond its
known working set, Z, however it assumes that Z is a metal for
minimization purposes. The parameters used for the Z atom in the
original force fields assume that it is an alkali metal.
problem:
Inorganic structures can be worked with in only a limited manner.
Usually, you can just visualize the structure and not do anything else with the
data.
Characteristic:
This software has a limited ability to work with metal atoms. It
has a generic metal symbol of Z which it uses for any and all metal
atoms it encounters.
problem:
When more than one metal atom appears in the same structure they
both have the symbol Z and can not be told apart.
problem:
The generic metal atom has been given generic values to be
included in the force field. These values correspond to that of an alkali
metal and may not fit the user's needs for other metal atoms.
Feature: Bonding Considerations
Characteristic:
The program allows the creation of dummy bonds between atoms to
allow you to tack portions of a complex structure together for different
purposes.
problem:
The software will only allow 6 bonds for any given atom. This
causes problems when working with metal complexes which may have more than 6
bonds.
Characteristic:
Bonding distances are automatic when reading in PDB files
for component atoms. The software ignores the connect portions of the file to
save space in its internal files.
problem:
Some bonds between atoms are automatically changed to undesired
preset distances. A file containing alpha carbons from a PDB file will
be shrunk from the actual distances to 1/4 those values if they are connected,
resulting in cute, but inaccurate alpha carbon traces. This also occurs when
the generic metal atoms are connected to other atoms in a structure.
problem:
Sometimes the bond lengths found in a PDB file are beyond the
normal limits for such a bond and weird connections are drawn in these
locations by the program. It may be necessary to check converted structures
for abnormal regions if anything looks unusual with the structure or it behaves
strangely in the software. Any missing bonds will have to be supplied by the
user.
Feature: Creating Backbones of Proteins
Characteristic:
You can automatically extract the backbone of a protein from a
displayed structure and work with it on the screen. This is an easy way to
reduce the amount of data on the screen to make secondary structure elements
more obvious.
problem:
To do this, the software assumes that a PDB or protein file has
been entered in a given manner. It expects that the order of atoms for a
peptide residue is N, CA, C and O followed by the
rest of the atoms. When this is not the case, strange things are displayed on
the screen.
problem:
Included substrate or non-protein molecules are not kept in the
resulting structure. If these groups are needed for analysis, you must use
other means to get them.
Feature: Energy Minimization
Characteristic:
Structures can be minimized by using one of three supported force
fields. MM2 is the force field of choice for organic molecules,
OPLSA is preferred for proteins. AMBER can be used on anything.
At times a procedure in which a molecule is first worked on by one force field
and then polished up by a different one can result in faster minimizations with
the end result having lower energy values than would be possible by using of
just one force field.
problem:
The different force fields require different conditions,
MM2 needs hydrogens, while AMBER will crash if they are there.
Going back and forth can be confusing to new users.
problem:
Some atoms are more difficult to minimize than others due to the number
of possible states they may exist in. If the selected state is not in a chosen
force field, the minimization will not run. Nitrogen functional groups
can be hard to minimize because of this.
Feature: Growing Structures
Characteristic:
MacroModel allows you to automatically grow structures from a
library of pre-existing fragments. This can greatly reduce the amount of time
needed to both enter and minimize a structure. Organic molecules, sugars,
nucleotides and proteins can all be entered this way.
problem:
Proteins have a number of preset conformations. A random option is not
one of them. This can cause real problems unless this is known and areas
assumed to be random are handled in a special manner.
problem:
The default conformations of the program are not always obvious
to the beginning user and can cause problems. Structures are entered as sheets
without the user realizing it.
problem:
These preset conformations cause problems when minimization is
done on a protein. You must modify the input file to allow movement in the
backbone, or the process will be meaningless.
Feature: Modification Potential
Characteristic:
If you don't find what you need in this program, you can write
your own software to interface with it to supply the missing functions. A user
interface is supplied with the software to serve as a starting point for such
efforts.
problem:
This works well if all of your new code is stand alone. If you need to
call on existing MacroModel functions there can be problems finding out
exactly what subroutines are needed to do what you want, and finding out how
they work.
Feature: Size Limitations
Characteristic:
MacroModel can show 10,000 atoms in display and work with 1,000
atoms when minimizing. There are also other limitations as to the number of
atoms to be numbered on the screen at one time, how many alternate locations
are allowed, and how many atoms can be used in certain types of minimization
processes.
problem:
PDB data files now produce files that contain more than 10,000
atoms on a regular basis, which means that they need to be cut back in order to
be used properly in the program.
problem:
The use of subsets of the data may require decisions about what
is important and what is not, and require numerous refinements of the data in
order to get a usable data set for analysis.
It may appear from this listing of possible problems that the program
MacroModel is too full of problems to be of any use. All modelling
software has limitations. Learning to effectively use any piece of software
and being fully aware of its shortcomings allows a user to be more confident
about any achieved results and to appreciate possible errors. Assuming that
everything the computer produces is correct is an invalid assumption and you
must be able to assess your output's worth.
Modelling small peptides presents interesting problems. The use of the grow button from the MacroModel program results in structures which have their conformational parameters set to be a helix, sheet or turn. This prevents the minimization process from finding the lowest energy state for any structure created this way. One possible way around this problem is to create files of the necessary individual amino acids for a given small peptide, make the necessary bonds between the component residues in their proper order and then minimize the resulting structure.
This minimization takes a long time and the process must be checked to insure that it runs as expected. Finding energy saddles and other quirks in the process can produce faulty files and the process will need to be restarted.
The result of regular minimization runs is the conformation of the structure as it would exist in a vacuum. This is not the real world situation. To find a more realistic conformation, solvent effects must be taken into account. This is done by repeating the minimization process, this time using a solvent parameter set to modify the vacuum results. Different solvents can be chosen to reflect philic or phobic environments. The use of an existing output file that has been minimized is preferred to cut down on the time required to complete the solvent run.
Week 6 Exercise
This exercise will acquaint you with potential problems associated with the use of structural data. Being able to recognize and correct data problems improves the quality of your final results. Customization of output produces more readily understood results that help get points across to an audience.
Most of your work this week will be on model1. You will be learning about the strengths and weaknesses of molecular modelling software and the data it uses. While VMS is different from UNIX, the basic skills you have acquired so far will help you on this platform with a different operating system. However, you will be filling out the report form and creating images on ribozyme. So you will set up that working environment first and then move on to model1.
1) Activate the computer and log into ribozyme.
Pressing any key changes the terminal from screen saver mode to active. Move the arrow to the RIBOZYME icon in the Launcher window and press the mouse button twice. Successful connection to ribozyme is denoted by the appearance of a ribozyme information line and a login: prompt.
Once you are logged into your account, create a subdirectory for this week's work and copy over the necessary files.
% mkdir week6 % cd week6 % cp $GRAD_DIR/week6m/*.* .
Since you will be running MacroModel sessions on model1, log off of ribozyme and use the Launcher window MODEL1 icon to start a session on that machine.
2) Log into model1.
From the Launcher window, select the MODEL1 icon and press the mouse button twice. Successful connection to model1 is denoted by the appearance of a model1 information line and a Username: prompt.
Welcome to OpenVMS V6.1 Username:
Once the Username: prompt appears, log on to the machine by entering the same account name you are using on ribozyme, and then your password to the Password: prompt.
3) Advanced data entry of functional groups.
To further explore the use of MacroModel to enter functional groups, get into the program using the instructions given below. In this section of the exercise, you will work with two nitrogen functional groups, the amine, -NH2, and the nitro, -NO2.
a) Activate MacroModel by typing mmv30.
b) Once in the program, respond to the first question by pressing RETURN. Answer the question about what terminal you are using by entering 7 for a Tektronix/Versa Term Pro.
c) When the working window comes up, move the cursor to the GROW button and press the mouse button. After the GROW button has turned green, use the cursor to select CH4 and press the mouse button twice, thus growing an ethyl group on the screen.
d) Move the cursor to NH3 and press the mouse button to cap off the molecule with an amine group.
e) Move the cursor to the GROW button again and deactivate it by pressing the mouse button. The purple box on the molecule denoting the future growth point will disappear.
f) Minimize the structure by doing the following steps: select ENERGY, select MM2 and when the cursor reappears, select Start. Depending on the accuracy of the initial structure, you may be asked to continue the minimization process. If asked, respond with y.
The MM2 force field has no problems with amine groups. It sometimes does with other nitrogen functional groups, such as nitro groups. Using the molecule on the screen, convert it to a nitro compound in the following manner:
a) Delete the lone pair of electrons shown on the nitrogen by selecting INPUT, then DELT and pressing the mouse button once. Move the cursor over to the : showing on the nitrogen. Look carefully to find it. If the : is found properly, the terminal will beep when the mouse button is pressed.
b) Convert the hydrogens on the nitrogen to oxygens by selecting the O from the side element options, and then moving over to the locations you need to change and pressing the mouse button again.
c) Draw a double bond between one of the oxygens and the nitrogen by selecting DRAW, moving to the chosen oxygen and pressing the mouse button. A beep should sound. Next move to the nitrogen and press the mouse button again, another beep followed by the appearance of a double bond should happen if this was done correctly.
d) Put a + on the nitrogen and a - on the oxygen not involved in the double bond by selecting the appropriate symbol from the side element options area and then moving over to the affected atom and pressing the mouse button.
e) Once this has been done, remove all of the hydrogens from the structure by selecting H DEL three times. The AMBER force field you will use for this minimization does not like hydrogens.
f) Minimize the structure by doing the following steps: select ENERGY, select AMBER and when the cursor reappears, select Start. Depending on the accuracy of the initial structure, you may be asked to continue the minimization process. If asked, respond with y. Record the final energy below.
energy value _________________________________________________________
g) When the minimization process is complete, select INPUT and then H ADD three times to add the hydrogens back on. The MM2 force field can handle the nitro group, however it may get hung in energy saddles and take a long time to arrive at a final result.
h) Now re-minimize the structure using MM2. Select ENERGY, select MM2 and when the cursor reappears, select Start. If asked to continue, respond with y. Record the final energy value on the next page.
energy value _________________________________________________________
4) Growing a protein.
Enter the first 10 residues of the protein melittin. This protein is the active ingredient that produces the swelling resulting from a bee sting.
gly-ile-gly-ala-val-leu-lys-val-leu-thr
Following the instructions below, enter this sequence first as a helix and then a sheet.
a) Select INPUT to get into the data input mode. For protein growth, select PEPTID. Once this screen appears, clear off the previous structure by selecting DELT three times and confirming the complete deletion by entering y. With a clear screen, select GROW. Since the fragment is first to be studied as a helix, select Helix from the side options. Move the cursor to select the individual amino acids in the order they are given above. As each one is chosen, it will appear on the screen in the growing peptide chain.
At this point you can write this structure to a file or simply go on to the next part of the growing process. If you want to write out this file, select Write, enter a filename for the data, and give a structure name for the data file. The cursor returns when the process is completed.
b) Having seen this fragment as a helical structure, redraw it in a sheet conformation. Select DELT three times, and confirm the deletion by responding with y to the question asked. Then select Sheet, because the Helix button is still active, and change the mode of peptide entry. Once the Sheet button is green, select the individual amino acids as before. Notice the difference in the actual amount of the physical space each conformation occupies.
Note: Any structure entered this way will keep its entered backbone configuration in the minimization process. If you truly want to wiggle a protein, you need to change the minimization parameters to allow the backbone as well as the side chains to be mobile, or create the protein in another manner.
Get out of the program by selecting STOP and respond with y to the Confirm Program Stop (Y/N):, your screen goes white. Select the Emulation from the Versa Term Pro menu bar and select DEC VT220 from the menu presented. The screen changes color from white to blue and has the question in it, Delete the current log file (Y/N):. Respond with y. The y you try to enter turns into ù and there is a message of junk on the line below the prompt. Select Emulation from the Versa Term Pro menu bar again, this time selecting Reset Terminal from the menu options presented. You can now continue with your computing tasks.
5) Converting data files to be used with MacroModel.
When working in the modelling area, it is always necessary to have the latest data to work with. At times this means contacting the lab actually doing the crystallization work on your substance of interest for their latest results. If you are lucky, they may even decide to share these with you. What you will receive from them is their own version of a PDB formatted output containing the desired information. You have no idea how this file was actually created. It may be the result of their software for determining the crystal structure. It could have been produced with a text editor. It might even be hard copy of the data. Whatever form it is, you need to get it into the computer in a form you can work with.
Three data files have been given the logical names PDB1, PDB2 and PDB3. Each of these files has something wrong with it. It could be as simple as not having the MASTER and/or END line at the end of the file to other problems that affect how the data is displayed on the screen in certain situations.
Determine what is amiss with each of these mini PDB files by first copying them over to your account. Use the examples given as guides. Note the actual names of the files as they are copied over into your account.
$ copy pdb1 *.* [real name _______________________________] $ copy pdb2 *.* [real name _______________________________] $ copy pdb3 *.* [real name _______________________________]
Use the following approach to get these files converted into a MacroModel usable format:
a) Check for the obvious problem of not having the proper ending lines in the file. This can be done by just displaying them off onto the screen with the type command. If any of the files are lacking these lines, go into the eve editor and add them.
$ type pdb1.data
b) Check to see if the program, bfiler, will work with the data. If any problems are reported back by the program, start checking to see if there are any tabs in the data. You can't spot this problem visually. Use the error message to help pin down the location of the problem. To locate a tabbing error, press FIND followed by the TAB and RETURN keys. The terminal will report if any tabs were found. Move to the next real character beyond the tab and use delete to remove the tab. Fill the gap by using the mouse button to move the cursor back to a position where the information in the line is properly aligned with the rest of the data file.
There will be times when tabbing problems will be scattered through the file, sometimes more than one error to a line. There may be many of these errors. Keep removing them until the data file can be successfully converted.
c) Get into the MacroModel program and check to see if the components of each residue are given in the correct order for the backbone feature of the program to work. This is done by reading in the converted file, which by the way now has the extension, bdt. Select READ, enter the filename including the extension, and press RETURN for the response to the structure number prompt. Selecting ANALYZ, SETS, MainS, DISPLA, and Dis will show if any problem exists with protein data. If an error message appears after selection of MainS there is a definite problem with the ordering of the atoms in the file. If unconnected atoms appear or only part of the backbone is displayed after Dis has been selected, there are also problems with the file.
Sometimes when working with partial data, a data file may not start with a nitrogen and therefore the software will not recognize it as a protein. If you find this to be the case, delete the partially given residue from the ascii version of the file and use bfiler again. You can't create atomic coordinates, only work with what you are given. If non-standard ordering of atoms occurs, use the program pdb_fix to correct the situation. If a problem still occurs with a file, use the NUM button to number the atoms and try to determine what is wrong with that atom's coordinates. This is done by converting the MacroModel file into a PDB formatted file by using mmodpdb and searching the line containing the data for the problem atom to find the error. Some numbers can easily be substituted for one another if the data was entered by hand such as numbers 6 and 9.
To insure that you have a feel for what a properly set up bare bones PDB file should look like, a file known by the logical name of PDB_pattern has been created. Copy this file over to use for comparison purposes. A bare bones PDB file just needs the coordinate data followed by a line with just MASTER and another line with just END in order for the bfiler program to work. Study this file before proceeding.
$ copy pdb_pattern *.*
The following examples are given to help you with the use of the programs bfiler, pdb_fix and mmodpdb. Refer to them for guidance in using the software.
Try this example for using the program bfiler. User input shown in bold type. In the example, lastname.out represents the name of the data file being used.
$ bfiler
BFiler (v 0.2)
27-NOV-96 08:37:18
BFiler: SELECT A MENU ITEM FROM BELOW--
HELP=Information
TAPE=Read Brookhaven format files Brookhaven tape and
translate to MMOD format,
COPY=Copy files from Brookhaven tape to disk
without translation
DISK=Translate Brookhaven format files to MMOD files
BARE=Translate Bare Brookhaven atom table (from file(s)
disk) to MMOD format file(s)
EXIT=Exit BFiler
BFiler> disk <rtn>
BFILER-DISK:This routine attempts to translate Brookhaven
format files which are on a disk
BFILER-DISK:Continue?(y): <rtn>
Default suffix is ".BRK"
Type in the names of the files you want to process,
Hit return after each code name and<
a bare "." to finish>
lastname.out <rtn>
.
Below is a list of names for files you want to translate --
Options: (1) type in corrected entry;
(2) type "i" to insert an entry,
(3) type "x" to delete entry,
(4) type "." to finish,
(5) hit return to verify entry:
lastname.out <rtn>
Go back and re-edit the filecodes?(n)> <rtn>
Looking for file lastname.out
Reading lastname.out
Reading atomic coordinates...
Typing atoms...
Creating bond entries...
BFiler: SELECT A MENU ITEM FROM BELOW--
HELP=Information
TAPE=Read Brookhaven format files Brookhaven tape and
translate to MMOD format,
COPY=Copy files from Brookhaven tape to disk
without translation
DISK=Translate Brookhaven format files to MMOD files
BARE=Translate Bare Brookhaven atom table (from file(s)
disk) to MMOD format file(s)
EXIT=Exit BFiler
BFiler> exit <rtn>
An example of using the program pdb_fix. User input shown in bold type. In the example, pdb1.try represents the name of the desired input file and pdb1.out the name of the created output file.
$ pdb_fix Program PDB_fix This program converts nonstandard PDB files into ones that will work in MacroModel Enter name of file to work with: pdb1.try <rtn> Enter name of output file created: pdb1.out <rtn>
If a data file was used that contained HETATM lines, comments would appear showing that such lines had been found and that the labels for them were unknown. Check the end of this convert file to insure that it has the necessary MASTER and END lines for bfiler conversion.
An example for using the program mmodpdb. User input shown in bold type. In the example, phenol.dat represents the name of the desired input file and phenol.pdb the name of the created output file.
$ mmodpdb THIS PROGRAM READS V1.5-2.0 MACROMODEL STRUCTURE FILES AND PRODUCES FORMATTED PDB STYLE OUTPUT FILES Enter MacroModel input filename: phenol.dat <rtn> Enter .PDB output filename: phenol.pdb <rtn> Charge file (.CHG) not found, charges set to 0.0 [structure name if any] Enter MacroModel input filename: <rtn> FORTRAN STOP
Look at each one of the PDB files, determine what is wrong with each file and then correct it. This may require going through the checking process a number of times until all the errors are removed and the data files will work correctly. Record below the nature of the errors with each file and how you corrected the situation.
PDB1 errors:
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
PDB2 errors:
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
PDB3 errors:
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
When you have fixed the files and run them through bfiler and display
them in MacroModel to be sure that there aren't any other problems with
them. Use this condensed instruction set to do the task.
a) Activate MacroModel by typing mmv30.
b) Once in the program, respond to the first question by pressing RETURN. Answer the question about what terminal you are using by entering 7 for a Tektronix/Versa Term Pro.
c) Select READ. Enter the name of the file to be displayed (remember the bdt extension). Respond with pressing the RETURN key to the Structure Number: prompt. Clear off the former image if necessary. Look closely at the presented image.
d) Check to see that the data is recognized as being from a protein by selecting ANALYZ, SETS, MainS, DISPLA, Dis. If this doesn't work or only a small fragment is shown, go back and work on the data file until it does work. [You may need to trim off atom lines that doesn't let the data start with nitrogen line of a residue.]
e) Rotate the structure to see if there are any data points that are way out of line. If so select Clip to zero in on the region with the errand point. Then from the ANALYZ , select NUM to have the atom numbers displayed on the screen. Armed with this information, go back and see if there was a confused number in the original data set.
f) Get out of the program by selecting STOP and respond with y to the Confirm Program Stop (Y/N):. Select the Emulation from the Versa Term Pro menu bar and select DEC VT220. To the Delete the current log file (Y/N):. prompt respond with y. Select Emulation from the Versa Term Pro menu bar again, selecting Reset Terminal from the options presented.
6) Working with actual data from a server.
Remember in exercise 4 (see pages 15 - 18) when you worked with an actual data file captured via GOPHER from the PDB server? A file that was brought over in a similar manner is what you will be working with next. Copy over the logically named file hydrogen to your account twice to have copies of the data to use. The first copy you will leave as it is for reference and the second you will modify (remove the extra hydrogens).
$ copy hydrogen *.* $ copy hydrogen hydrogen2.txt
This file has been cleared of any mailer information. Use the eve editor to reduce the hydrogen2.txt file to a bare bones file by removing all the text prior to the actual ATOM lines.
$ eve hydrogen2.txt
Now look at the actual data. Note that there are extra hydrogens listed in the data set. An example of what is meant by this is given below. Remove these extra hydrogen sections from the data set. Remember to keep important hydrogens like the hydroxyl hydrogen in THR and SER and the hydrogens connected to the terminal nitrogens in ARG and LYS.
ATOM 9 1H ASP 1 0.162 -1.467 -27.365 1.00 7.00 1ATF 53 ATOM 10 2H ASP 1 -0.499 -2.002 -25.976 1.00 7.00 1ATF 54 ATOM 11 HA ASP 1 1.894 -2.151 -25.180 1.00 7.00 1ATF 55 ATOM 12 1HB ASP 1 3.262 -0.264 -26.086 1.00 7.00 1ATF 56 ATOM 13 2HB ASP 1 2.918 -1.481 -27.300 1.00 7.00 1ATF 57 ATOM 14 HD2 ASP 1 2.223 2.077 -28.373 1.00 7.00 1ATF 58
When finished with this operation, convert this data file and hydrogen.txt into a MacroModel-compatible files with bfiler. Then use the following condensed instruction set to look at the resulting files.
a) Activate MacroModel by typing mmv30.
b) Once in the program, respond to the first question by pressing RETURN. Answer the question about what terminal you are using by entering 7 for a Tektronix/Versa Term Pro.
c) Select READ. Enter the name of the file to be displayed, hydrogen.bdt. Respond with pressing the RETURN key to the Structure Number: prompt. Look closely at the presented image.
d) Select READ again. Enter the name of the file you remove the extra hydrogens from, hydrogen2.bdt. Respond with pressing the RETURN key to the Structure Number: prompt. Do not delete the current structure. Look closely at the two images. Record below your impression of them.
impressions:
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
e) Get out of the program by selecting STOP and respond with y to
the Confirm Program Stop (Y/N):. Select the Emulation from the
Versa Term Pro menu bar and select DEC VT220. To the Delete the
current log file (Y/N):. prompt respond with y. Select
Emulation from the Versa Term Pro menu bar again, selecting Reset
Terminal from the options presented.
7) Customizing a data file.
There will be times when customizing a data file will make the information it contains easier to understand, especially when you are trying to emphasize one portion of a complex structure. In a project done for Dr. Wm. Trumble (U of I) on cardiotoxin, the emphasis of the grant request was mutation studies on the protein. To illustrate this point, a file was created showing the backbone of the protein for all amino acids not to be changed, and having a full representation for the mutation points and the disulfide bridges. These were color coded to show the different mutation points (red, blue and purple) and the location of the bridges(aqua).
To see what is meant by this verbal description and just how effective it can be, read in the following files on the same screen and compare them for yourself. You should be familiar enough with the program by now that you don't need exact instructions to find the buttons you need to do simple tasks within the program.
Start up the MacroModel program. Get into the analysis mode by selecting the ANALYZ button. Reduce the amount of information to be displayed on the screen by selecting A LAB so it won't show atom labels. Read in one file by selecting READ, respond with cardio for the file, press RETURN for the structure number. Once the file has been completely read in, pull up the second one by selecting READ, giving mod_cardio for the file, press RETURN for the structure number and this time respond with n to keep the current structure. Now on the screen, side by side, are the cardiotoxin-modelled structure in its entirety and the modified version of this file coded to emphasize the points of interest. Record your observations below. Exit the program.
observations:
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
Now create a similar type of file yourself using your working version of the
PDB3 data. What you will want to display is the type of interactions possible
between the charged amino acids of this fragment. There are five of them in
this file, ASP, GLU, ARG, HIS and LYS. For the non charged amino acids, show
only the backbone atoms.
MacroModel can color residues by charge by first reading in a file, going into the ANALYZ mode, and selecting DISPLA, followed by RChrg. Acidic residues will be colored red and basic ones dark blue. The non charged residues are yellow.
Creating this type of file requires combining MacroModel functions, i.e., writing of color coded structures and editing skills. Use the PDB ascii version of this file that you got to work in bfiler. Do two operations on this data, first strip that data down to its backbone components via the program backbone. Then with the original data, do a search to get the complete coordinates for the five residues of interest, ASP, GLU, ARG, HIS and LYS. Since backbone is a newly introduced program, an example of its usage is given below. User input is shown in bold type. In the example, pdb3.out represents the name of the input file and pdb3.out2 that of the output file.
$ backbone Program backbone This program pulls the backbone data from PDB files Enter name of file to work with: pdb3.out <rtn> Enter name of output file created: pdb3.out2 <rtn>
The VAX sea command is similar to grep in UNIX. In the command line given below, the desired file is searched for the terms asp, glu, lys, his and arg. The results are written to the output file res.lis. In this example command line PDB3.data represent the file to be searched.
$ sea/out=res.lis PDB3.data asp,glu,lys,his,arg
By appending the results of the search for the residues of interest to the results of the backbone file, a composite file is created. It just needs to have its parts rearranged and some lines deleted in order to have all the necessary information for the creation of the needed data file. In the example given below xxxxx.xxx represents the name of the PDB3 backbone file you created.
$ append res.lis xxxxx.xxx
Now, use eve to get the file into its final form. You need to move the complete data for the five residues of interest into their proper places and delete those lines that appear twice. Correct any residue numbering problems that may exist. Then convert the file into a MacroModel compatible form with bfiler.
The final step is to get into MacroModel, select the ANALYZ mode, A LAB, DISPLA and RChrg. The data is now shown with complete coordinate data for the charged residues and they are color coded as to the type of charge they contain. Non-charged residues are only shown by their backbone atoms and are yellow. Write this data to a file called rchrg.
8) Color coding protein backbones.
The color coding aspects of MacroModel can be used to set apart one protein chain from another. This feature coupled with the stripping of a protein down to its backbone members can greatly increase the ease with which the secondary structure of complex proteins can be looked at and studied.
To try out this technique, select the ANALYZ mode, A LAB and then READ in the structure of alpha chymotrypsin A, which has the PDB access code of 5CHA. This logical name has already been defined upon logging into your account and it can be entered directly into the modelling program. Strip the protein down to its backbone, using SETS, MainS, DISPLA, and Dis.
This protein is very complex and has two chains. Finding the two chains is difficult in the current method of displaying the data. Select Mol, then move to some point on the molecule and press the mouse button. If you were over a coordinate location, the terminal would return a beep. If you get no beep, move off to another spot and try again. Next select Mono, enter a color, one letter will do. There are 16 color choices and the common ones are present in the program. For this pass, use a for aqua. Respond to the question about which atoms are to be colored by entering w for working set. The affected chain will be erased from the screen and then redrawn in the desired color. After the colored chain is completely redrawn, select Mol again, and this time move over to a non-aqua colored atom and press the mouse button. Go through the color selection process again, making this new chain yellow. It may be necessary to select the Updat button to get a good picture of this color coded molecule once all the chains have been colored. There is a break in each chain at its beginning which you can spot if you study the redrawing process closely.
Write out this data into a file called m5cha. To the Save displayed fragment only (Y/N): prompt, respond with y.
9) Creating a small peptide model.
[These instructions assume that you are in the ANALYZ mode of MacroModel.] Put the atom labels back on by selecting A LAB. READ in the files with the logical names p_glu, his and pro into the working area after having deleted the previously created data from the screen. These three amino acids are the components for the TRH protein. Move over to the input portion of the program by selecting INPUT. Create the necessary bonds between the carbonyl groups of the various residues and its adjacent amino group by using DRAW. Put a hydroxyl group on the proline residue to cap off the peptide again using the DRAW function. Check the structure for any missing or additional hydrogens by using the H DEL and the H ADD buttons. Check for nitrogens with four bonds shown and correct this situation if found by using DELT and removing offending hydrogens. Go into the energy portion of MacroModel by selecting ENERGY, change the number of iterations to 5 by selecting the It/S button. Use the MM2 force field for the minimization process. Select Start to start the actual process. If no problems pop up as this process is run, write the results of this minimization to a file with the name trh1.
Go through this process again, only this time after reading in the initial three residues from the INPUT mode, move over to the ORIENT portion of the program. At this point use the Mol and the Rot Z buttons to change the orientation of the three residues so that the p_glu is the first part of an inverted U, the his is at the top and pro the right side. Move back to INPUT, ORGANI section by selecting ORGANI. Make the necessary bonds between the carbonyls and their adjacent amine groups and cap off the structure with an hydroxyl on the proline. Check the structure as before for 4 bonded nitrogens and too many hydrogens on a given atom. Do a 5 iteration minimization on the resulting structure and if successful write out the results of this process to a file, giving it the name trh2.
Get out of the program. Copy over to your account the logically named file, batchlesson30, once for creating the trh1 batch job and once for the trh2 job. Rename these files to reflect your last name and the molecule you are working with. Then edit the batch files with eve to reflect the actual data. A copy of the batchlesson30 file is given below. Change the set def line (put and ! symbol after the $ and the aspirin and aspirin_out lines to contain the name of the input file and the desired output filename. In the example lines the xxx represents the first three letters of your last name.
$ copy batchlesson30 xxx-trh1.com $ copy batchlesson30 xxx-trh2.com $! pcd xxxxx xxxxx xxxxx xxxxx ffff.ffff ffff.ffff ffff.ffff ffff.ffff $ set def [xxxxxx] $ define incloc1 disk2:[public.mmv30.mmv30.inc1] $ run disk1:[manage.loging]batchm30 $ run incloc1:batchmin.exe aspirin aspirin_out DEMX 0 20.0000 FFLD 1 BGIN READ 1 MINI 2 1 9945 CONV 2 1 END
When the editing is completed, submit your batch jobs in the following manner, where xxxxx denotes the name of your batch job com file:
$ batchs xxxxx
You will need to wait until the results of these minimizations come back to finish up this section of the exercise. Once the two batch jobs are finished look at the results in MacroModel. Are the two structures similar? Choose one of the two output files to continue to explore with.
Normal minimizations create conformations of molecules as they would exist in a vacuum. You can run additional minimization runs that take into account the effect of solvents on the molecule. To do this copy over the logical file solvents. A copy of this file is given below. It is very similar to the regular batch command file for a MacroModel run, except it contains a line that allows you to select a solvent. A 3 in the first place of the SOLV line picks the best solvent model to use for solvent runs. The second number in that line picks the solvent to be used, 1 - water, 4 - MECL2 and 5 - CHCL3.
! pcd xxxxx xxxxx xxxxx xxxxx ffff.ffff ffff.ffff ffff.ffff ffff.ffff $ set def [xxxxxx] $ define incloc1 disk2:[public.mmv30.mmv30.inc1] $ run disk1:[manage.loging]batchm30 $ run incloc1:batchmin.exe aspirin aspirin_out DEMX 0 20.0000 FFLD 1 SOLV 3 1 BGIN READ 1 MINI 2 1 9945 CONV 2 1 END
Copy the solvents file twice into your account, once to control the running of a solvent run using water and secondly for a CHCL3 solvent run. Use appropriate extensions to tell them apart. Copy the output file from the trh minimizations that you have chosen to use twice as well. Once for the water run and another time for the CHCL3 run. MacroModel batch minimization runs lock the data file they are using and so a separate copy of the data set must be available for the runs in case they are both running on the platform at the same time. Make the necessary modifications to the two solvent command files with the eve editor and send them off to the batch queue as you did with the initial two batch jobs. An example of doing this is given on the previous page. The xxx in the example lines is the first three letters of your last name. The trhx_out.dat in the example lines represents your selected trh minimized output file.
$ copy solvents xxx-trh-water.com $ copy solvents xxx-trh-chcl3.com $ copy trhx_out.dat trh-water.dat $ copy trhx_out.dat trh-chcl3.dat $ batchs xxx-trh-water $ batchs xxx-trh-chcl3
When these results come back, compare them to the original minimization results. Create a comparison file containing the original minimized structure, the water result and the CHCL3 result. This is done by getting into MacroModel and reading in the original file, then the water result and finally the CHCL3 result.
The image on the screen is fine in MacroModel, but the molecules are too close together for a Molscript image and extra bonds will be drawn between the three structures. To prove this for yourself, select ANALYZ, ADist and determine some distances between the atoms on the right side of the original structure and the left side of water minimized one. Do likewise for the close atoms on the water and CHCL3 minimized structures. Record these distances below.
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
The distances you have recorded are below the default distance in the
Molscript program for a bond (1.9 angstroms) and therefore would have
caused problems with the image. To prevent this from happening, do the
following steps to produce the three structures aligned in a column instead of
a row with enough distance between them to provide a distinct image of each
one.
Now select INPUT, ORIENT, and Mol and use the cursor to select an atom from the original minimized structure. Pick Trans, enter y to the prompt and enter 12 for the distance. The molecule should move up to overlap the top edge of the working window. Enter * to get stop moving the structure in the y direction. Back at the direction prompt again, enter x and at the distance prompt 12.5. This should move the structure directly above the one in the middle of the working window. Press the RETURN key to get out of this translation process and back to a regular cursor again.
It is time to move the molecule on the far right. Select Mol, and pick an atom on that structure. Select Trans, respond with y and to the distance prompt with -12 this time. Again the molecule moves to overlap an edge of the working window, this time the bottom one. Enter * to get stop moving the structure in the y direction. Back at the direction prompt again, enter x and at the distance prompt -12.5. This should move the structure directly below the one in the middle of the working window. Press the RETURN key to get out of this translation process and back to a regular cursor again.
Select the Scale button and move the cursor to first the bottom left hand corner of the working window and press the mouse button getting a red cross and then the upper right hand corner and pressing the mouse button. This re-scales the image to fit better within the working window.
The data in its present form is still suffering from hidden problems. They all arise from the default bond distance. If two hydrogens on the same carbon are too close together a bond is drawn between them. A close up view of the problems in your data set is given in the image below. This image is the result of running Molscript on the data as it now stands. The points of concern are labelled from one to six. The first five are easy to fix, the sixth is a little harder.
Normally in this situation, you just want to extend the length of the bond a little to get past the 1.9 angstrom default distance. One quick and dirty way of doing this is to just replace the offending hydrogen with another atom. Fluorines are a good choice for this since they rarely are present in normal protein structures. To do this select ORGANI and then the F from the selection of atoms given there. With the F now colored green, move the cursor to each of the first five positions on the structure exactly at the end of each line and press the mouse button. A red F should appear and the bond color changed from all green to half red and green.
Position six requires another approach. Select DRAW and move the cursor to a spot on the working window a little above the hydrogen in position 6. Press the mouse button, a C appears on the screen. Then move the cursor to the oxygen that the original hydrogen is connected to and press the mouse button again. A green bond is now on the screen. Delete the original atom is position six.. Select H from the atoms listed on the side and then change the end of the new green bond into a hydrogen atom. A H appears on the screen. Now normalize the oxygen bond between the new hydrogen and the original oxygen by selecting Norml and moving the cursor to the oxygen and pressing the mouse button. The position of the hydrogen will move and be more aligned with the bond between the oxygen and the carbon atom of the carbonyl group. Now select a Br from the selection of atoms from the side of the screen and change the newly moved hydrogen into a bromine atom. A Br will appear on the screen and its color is a different shade of green than you are used to seeing in this program. In this particular case, a fluorine didn't create a long enough bond to solve the problem.
Apply these corrections to all three structures, then write the resulting modified data to a file called solv and exit from the program. You may find that clipping out one structure at a time to work on makes the process much easier to do.
Because working with all the data displayed on the screen at one time would be confusing and could cause errors, use only a small portion of the screen at a time. This is done by selecting the Clip button at the upper right-hand side of the screen. Clip allows you select a portion of the screen to work with and then expands that selected portion to fill the entire working area. Once Clip is on (shown in green), move the cursor to a point on the screen that will serve as the lower left-hand corner of the area you desire to work with. Once in position, pressing the mouse button will cause a red indicator to mark the spot. Now move the cursor to the desired upper right-hand corner and press the mouse button again. After the indicator is shown, the entire working area clears and the desired section of the molecule will fill the screen. To return to the original data set select Clip and press the mouse button twice.
10) Working with selected molecules.
Using your selected molecule, color code the backbone of the structure to display its charged residues. Use these instructions to accomplish this task. Start up MacroModel, and READ in your selected molecule's bdt file. Get into the ANALYZ mode, then select Sets, followed by MainS, DISPLA, Dis and then RChrg. You will be informed All atoms or working set may be colored Input A or W (A/W). Respond with w for working set to correctly display the desired information. Again acidic residues will be colored red and basic ones dark blue and non charged ones are yellow. Save your work to a file named selback. Exit the MacroModel program.
11) Converting files and moving them to ribozyme.
Use the mmodpdb program to convert the following files into PDB formatted files; rchrg.dat, m5cha.dat and solv.dat. Use the example on the next page to help you do this. Remember that you can cycle through the program a number of times with different data files before exiting it by pressing the RETURN key to the input filename prompt. In the example xxxxx.xxx represents the name of the input file (MacroModel formatted data file) and xxxxx.pdb represents the name of the converted output file.
$ mmodpdb THIS PROGRAM READS V1.5-2.0 MACROMODEL STRUCTURE FILES AND PRODUCES FORMATTED PDB STYLE OUTPUT FILES Enter MacroModel input filename: xxxxx.xxx <rtn> Enter .PDB output filename: xxxxx.pdb <rtn> Charge file (.CHG) not found, charges set to 0.0 [structure name if any was given] Enter MacroModel input filename:<rtn> FORTRAN STOP
What is going to be done with this data on ribozyme is to create some more Molscript images, therefore you will need to modify some of these files here where the editor is more powerful than on ribozyme.
The m5cha.pdb data files needs a lot of modifications. Since the data contains 2 chains of approximately 245 atoms each, using an editor that can do replacements is very handy. Use the following instructions to make these changes. First you need to extract out those coordinate data lines that correspond to CA lines. In the first chain these are the C02 line and in the second chain the C06 lines. Use the VAX sea utility to create output files containing these lines and then combine then with an append command. Rename the resulting file to whole.pdb and edit that file to replace the C02 and C06 terms with CAspace.
$ sea m5cha.pdb C02 /out=part1.pdb $ sea m5cha.pdb C06 /out=part2.pdb $ append part2.pdb part1.pdb $ ren part1.pdb whole.pdb $ eve whole.pdb
From all the work you did on the solv file getting it into shape, you would think that would be enough, but is not. With the data converted into a PDB formatted file, it is time to organize the data into units that will make its processing produce the desired results.
Use the eve editor to examine the data file. Notice how the data is grouped. There are no real obvious signs as to which coordinate lines belong to which of the three molecules. The data just seems to repeat itself three times. However if you look at the y coordinate column, you can see where the differences are. The first (top) molecule has y values from 10 to 18, the second (middle) from -1 to 6 and the third (bottom) from -6 to -13. Each of these regions start with two lines similar to that given below. The new bromine atoms for all three molecules are given at the end of the coordinate lines, just before the CONECT lines.
x y z ATOM 1 H01 GLU A 1 14.874 12.675 3.438 1.00 0.0000 0 ATOM 2 N02 GLU A 1 14.390 12.984 2.595 1.00 0.0000 0
Within each data set the information is organized in the same manner. First the backbone and side chain atoms are given for each amino acid, then the hydrogens to be attached to the amino acids. In our case some of these hydrogens are now fluorines, one each for GLU and HIS and four for PRO. The capping hydroxyl group atoms have the residue name UNK and have a chain number of zero. There are a total of 47 atoms in each structure.
Start out your modifications on this file by first removing all the CONECT lines at the bottom of the file. Molscript doesn't use them. Then collect all the F lines at the bottom of their respective section of the data. Change the amino acid name on these lines from that given to UNX for the first molecule, UNY for the second and UNZ the third. Change the chain numbers to 2 for the first molecule, 3 for the second and 4 for the third. Finally move the BR lines into their respective places from the bottom of the file to the end of the F lines for each molecule and change their amino acid code and chain number to match that of the rest of the F section they are now in.
To finish up this week's activities, you will need to put the converted PDB formatted files into your account on ribozyme. This is done by using FTP (File Transfer Protocol) to move the data. Instructions for using FTP on Model1 are given below.
$ ftp ribozyme.vadms.wsu.edu
When the ribozyme machine prompt appears, enter your account name and password on that computer. Replace the bcsxx of the example with our own account name. User input shown in bold type.
model1.vadms.wsu.edu MultiNet FTP user process 3.4(111) Connection opened (Assuming 8-bit connections) <ribozyme.vadms.wsu.edu FTP server ready. RIBOZYME.VADMS.WSU.EDU>l bcsxx<rtn> <Password required for prcadams. Password:(enter your own password<rtn>) <User prcadams logged in. RIBOZYME.VADMS.WSU.EDU>cd week6<rtn> <CWD command successful. RIBOZYME.VADMS.WSU.EDU> type ascii<rtn> Type: ASCII (Non-Print), Structure: File, Mode: Stream RIBOZYME.VADMS.WSU.EDU>put hydrogen.txt<rtn> To remote file:<rtn> <Opening ASCII mode data connection for 'hydrogen.txt'. <Transfer complete. RIBOZYME.VADMS.WSU.EDU>put hydrogen2.txt<rtn> To remote file:<rtn> <Opening ASCII mode data connection for 'hydrogen2.txt'. <Transfer complete. RIBOZYME.VADMS.WSU.EDU>put rchrg.pdb<rtn> To remote file:<rtn> <Opening ASCII mode data connection for 'rchrg.pdb'. <Transfer complete. RIBOZYME.VADMS.WSU.EDU>put whole.pdb<rtn> To remote file:<rtn> <Opening ASCII mode data connection for 'whole.pdb'. <Transfer complete. RIBOZYME.VADMS.WSU.EDU>put solv.pdb<rtn> To remote file:<rtn> <Opening ASCII mode data connection for 'solv.pdb'. <Transfer complete. RIBOZYME.VADMS.WSU.EDU>quit<rtn> <Goodbye.
This completes your work for the week on model1. Log off of the machine by doing the following command.
$ log
Log into ribozyme. From the Launcher window move to the RIBOZYME icon and press the mouse button twice. Successful connection to ribozyme is denoted by the appearance of a ribozyme information line and a login: prompt.
Once you are logged into your account, move over to the week6 subdirectory.
% cd week6
section 12 a - creating Molscript images.
Produce the desired Molscript images using the following instructions. Molscript uses a input control file to control the creation of the image. In this file is the name of the PDB formatted file to be used and the what is do be done to this data set. In the case of the five images that you will be creating, the control files are already in this location. Each of the five images will point out a different strength or weakness of the Molscript program or the creating of PDB formatted files from MacroModel data..
image #1: -
This image uses the hydrogen files you worked with on model1. These files are already in standard PDB format and so no conversion process was require. In the images drawn a comparison is made between a structure with a lot of hydrogens and that of one with minimal hydrogens. Shown at the top of the page is the structure with a lot of hydrogens, below that with minimal hydrogens. A ball-and-stick representation is used.
% molscript <hydro.in> (your lastname)-hydro.ps
Print off a copy of the results.
% lpr (your lastname)-hydro.ps
Look at the results and record your observations below.
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
image #2: -
This image uses the rchrg data. Here the desired task is to produce an image with a ribbon showing the general shape of the backbone of the peptide and the charged side chains shown. To do this task you need to modify the rchrg.pdb file. To draw the ribbon you will need to change the C02 in the residues to CA. Be sure to keep the spacing in the line correct. Then go through an remove the N01, C03 and O04 lines from all the residues given. You can delete the CONECT lines if you wish, Molscript only reads and acts upon information in a PDB file starting with ATOM or HETATM.
While you are editing the file notice that there is an X between the residue name and the residue number as in the example line given below. This X is known as the chain identifier. When there is a character in this position of a PDB coordinate data line that character must be used along with the residue number in Molscript secondary structure lines such as that given below.
ATOM 1 N01 ALA X 71 40.693 40.928 40.072 1.00 0.0000 0 helix from X71 to X82;% molscript <rchrg.in> (your lastname)-rchrg.ps
Print off a copy of the results.
% lpr (your lastname)-rchrg.ps
Look at the results and record your observations on the effectiveness of this image below.
____________________________________________________________________
____________________________________________________________________
____________________________________________________________________
image #3: -
This image uses the whole.pdb data. Looking at this data shows that there are two chains, A and B. You have already spent a great deal of time getting this file in shape on model1. On ribozyme, you will need to observe the Molscript run and see if there are any data problems.
% molscript <whole.in> (your lastname)-whole.ps
Did the Molscript trace look normal? Where there any gaps in the data that were revealed by breaks in the helix numbers shown? Look at the Molscript screen trace to find the gap regions. Record the number of gaps and their locations below.
% more whole.pdb number of A chain gaps: _______________________________________________ locations: ____________________________________________________________ number of B chain gaps: _______________________________________________ locations: ____________________________________________________________
% pico whole.in
Create another Molscript image from your revised whole.in file. Print it off to make sure that everything worked.
% molscript <whole.in> (your lastname)-whole.ps % lpr (your lastname)-whole.ps
This image results the solv.pdb data. Examine the solv.in file to see how the modifications you put in are going to be used. Since the fluorine and bromine atoms are pretending to be hydrogens, their atomradius and atomcolour values have to be reset to match that of a hydrogen. Labels have been inserted so you can tell the molecules apart. The image has also been moved to the left to make room for the labels. The resulting image is ball and stick. You have already spent a great deal of time getting this file in shape on model1. On ribozyme, you will need to observe the Molscript run and see if there are any problems.
% cat solv.in % molscript <solv.in> (your lastname)-solv.ps % lpr (your lastname)-solv.ps
% rcp (your lastname)-hydro.ps teacher@ribozyme:receive % rcp (your lastname)-rchrg.ps teacher@ribozyme:receive % rcp (your lastname)-whole.ps teacher@ribozyme:receive % rcp (your lastname)-solv.ps teacher@ribozyme:receive
Rename the report form to have your last name, and then go into the file with pico to fill it in. Finally send it over to the teacher account.
$ mv week6m.week6m (your lastname).week6m $ pico (your lastname).week6m $ rcp (your lastname).week6m teacher@ribozyme:receive
This concludes your computing session for this week. Log off the ribozyme, get out of the emulator and back to the Launcher window screen.
Per J. Kraulis, "MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures", Journal of Applied Crystallography (1991) vol 24, pp 946-950.