Part I provides basic operation instructions such as Turbo-Frodo's hardware and memory requirements, the various ways of interacting with the loaded molecule, and how to run Turbo-Frodo, among other necessary operation information. It also covers the various files that need to be prepared, such as the Heap File, before displaying a molecule on the screen.
CHAPTER 1
Operation
The Turbo-Frodo software program will run on Silicon Graphics workstations possibly working under IRIX 4.0 but more likely 5.2 and above. The program uses the RGB color mode and thus can work on an 8-bit graphic board. Nevertheless, for better usage convenience, it requires 2x12 bitplanes and a Z buffer in order to see not only nicer electron density maps but especially accurate and impressive representations of CPK and secondary structure.
Use the following UNIX order from the general directory TURBO_DIR (ex:/disk_usr/turbo)
This makes it possible to generate the Turbo-Frodo tree. Modify your .cshrc as follows by typing,
or execute the progam entitled EXECUTE-ME and follow the instructions.
Start the program by typing >turbo or >turbo -d (select a default heap name, i.e., test.heap) or >turbo -t (cancel the display of the Turbo-Frodo starting logo, which can be useful if working on a non-graphic terminal).
For better performance, Turbo-Frodo requires some swap memory. If you have some free disk space, you should therefore try adding some swap memory. For machines running under system 5, you can add swap memory interactively. The following procedure adds 150Mb of swap space to the system by using a file in the /swap directory:
To make this swap area permanent (that is, automatically added at boot time), add the following line to /etc/fstab:
/swap/swap1 swap swap pri=300
There are four ways of interacting with the loaded molecule: via the mouse, via the keyboard by using the Textport commands, via the pull-down menus, or via the button box (or pseudo-button box).
With the mouse one can either access the pull-down menus by clicking the right button or interact with the molecules on the screen by clicking the middle button. The left button comes into use in the atom-addition mode (see this specific function).
The following combinations of mouse buttons can also be used (Active, or clicked-on buttons, are indicated in boldface):
A new feature now exists that helps you walk along the density map and have the closest possible view. You can now center on a density point, in the same manner as you do on an atom, by clicking the left and middle buttons on the mouse. When selecting a point on the screen, the program searches for the highest of the density points found along the current slab and centers on it. This feature is especially useful to look at empty MIR density maps.
Textport commands consist of orders and options. An order needs a target, whereas an option does not. Both types of commands have to be validated with the ENTER key on your keyboard.
The pop-up menus are activated when the cursor is driven up to the top of the pop-up menu head. The submenus or the options can be accessed accordingly by dragging the mouse down, keeping the right button depressed. This will make the various options appear in yellow. To choose an option, drive the mouse along the menu until the option required is highlighted in yellow. Releasing the right button on the mouse will activate the option, which will then be highlighted in green. To turn off the option, go to it with the mouse and click the right button again. This will change the color of the option back to yellow and deactivate it.
The button box (or pseudo-button box) makes it possible to move the entire molecule, or part of it, on the screen, in all three dimensions, by performing translations and rotations. It is thus possible to set the distance for stereoscopic viewing. With the button box, one can also display only a section of the molecule and move this section along the Z-axis. If you do not have a button box, the Textport PSEUDO-DIALS option will display on the screen a set of cursors controlling the same options as the button box, via the right button on the mouse. The dial functions are summarized in the display option called HELP DIALS.
To run Turbo-Frodo, just type >turbo or >turbo -t if you are working on a non-graphic terminal. The turbo -t option may be useful if you want to prepare particular files such as the Heap File, or the files necessary for NMR purposes. This will set the program running.
If this procedure fails, and if you are on an appropriate terminal connected to an Iris workstation, ckeck the environment by typing >printenv. Check the value of the variable TERM, which should be set to iris-ansi. Otherwise, correct it by typing >setenv TERM = iris-ansi.
Turbo-Frodo will appear on your graphic screen or not, depending on whether you have started the program with or whitout the -t option, and a Textport window will open, prompting you to give the name of the Heap File. Enter a name. If the name you entered corresponds to an already existing Heap File, the program will wait for the next command. Otherwise it will ask you whether or not you want to create this file. Answer accordingly.
If your Heap File already contains a structure, it can be loaded by typing turbo >load "molecule"or turbo >load "#", # being the data base number of the molecule you want to load on the screen. The data base number may be obtained by typing turbo>list heap. Conversely, a molecule can be unloaded by typing turbo >unload "molecule" or turbo>unload "#". Once the molecule has been loaded, if you want to display it, simply type turbo>go.
CHAPTER 2
Necessary Files
Before displaying a molecule on the screen, you need to prepare some files. First of all, you have to prepare the data bank in which all your molecules are stored, the Heap File. The Heap File is an infinite file and is limited only by your disk space. It contains data on the molecular structure in a format readable by Turbo-Frodo.
To prepare the Heap file, you have to read an ASCII coordinate file about your molecule. This file may by written either in the Protein Data Bank format (PDB) or in the Wayne Hendrikson format (WH). Accordingly, type either turbo>read pdb or turbo>read wh.
The program will then prompt you to give the molecule in the Heap File a name. This name should not be longer than 8 digits, otherwise only the first 8 digits will be recorded. The contents of the Heap File can be listed by typing turbo>list heap. Whenever you want to save changes in a molecular entry made on the screen, you must save it in the Heap File.
The Heap File is not readable without Turbo-Frodo. Therefore, if you want to use this information with other programs, it is necessary to prepare a file in the PDB, WH, or XPLOR readable PDB formats by typing turbo>make pdb or turbo>make wh or turbo>make xpdb, as appropriate. The program will prompt you to give the name of the output file.
While working within the Heap File, you can duplicate any existing stucture by typing turbo>duplicate "filename". The program will then prompt you to give the new name. You should avoid giving the duplicated molecule the same name. Conversely a molecule can be deleted from the Heap File by typing turbo>delete "filename"or turbo>delete "#", # being the number attributed to the molecule in the Heap File.
The Connolly Surface File and the Spline Connolly Surface. These files are located in remote programs, not in Turbo-Frodo itself. These programs are supplied, however, along with Turbo-Frodo and are placed within the /ms subdirectory.
The Map File has the same format as the classical DSN6 Frodo map. The Map File can be obtained from the Fourrier output of various X-ray packages (CCP4, PROTEIN, and so on) after conversion using the Mappage program, which is provided with Turbo-Frodo. All Mappage files required are placed under the /map subdirectory.
Assuming you have already loaded a molecule, you can then load an X-ray file by typing turbo>load map. The program will prompt you to give the name of the input file and to define the contour levels, colors, and so on, which are now required.
The Bones File must first be calculated with the supplied program named bones_turbo, which is located under the $TURBO_DIR/map subdirectory. It accepts maps from either CCP4 or XPLOR softwares. After calculation, the output file must be entered in Turbo-Frodo with the LOAD BONES Textport order.
This file contains the NMR constraints deduced from a nOe set collected on a NOESY spectrum recorded with a molecule. The format of an NMR file is the same as that used by XPLOR for NMR structure determination and refinement. It contains the name of the two protons between which a nOe has been determined, the distance determinated after calibration of the NOESY spectrum, and the lower- and upper-tolerance values.
In the following example,
assign (resid 1 and name ha) (resid 32 and name hb*) 4.0 2 1.0
a nOe has been observed between the alpha proton of residue 1 and one beta proton of residue 32. Since the methylene protons were not unambiguously assigned, a wildcard is used (hb*). This corresponds to the use of pseudo-atoms in distance geometry calculations. After calibrating the spectrum, the upper limit was set at 4.0 . The lower-tolerance limit was set so that the lowest acceptable distance would be the Van der Waals distance. The upper-tolerance limit compatible with the use of a pseudo-atom has been set at 1.0 .
Assuming you have already loaded a molecule, you can then load an NMR file by typing turbo>load nmr. The program will prompt you to give the name of the input file and then list the atoms present in the nOe file but not in the molecular structure. This often arises when you want to check experimental constraints on a structure originating from a PDB, which does not give the protons. In this case, you will have to generate these hydrogens by using, for example, the XPLOR package. Upon typing go, the structure will appear on the screen with the constraints displayed in different colors, depending on the difference between the constraint given in the input file and the real interatomic distance.
This file contains a list of all the protons in your molecule, including the pseudo-atoms and the NMR chemical shifts coming from your experimental data or from the literature. It can be generated manually, but it will be automatically produced by the EASY NMR assignment software. The format is that used by EASY:
128 4.123 0.002 0.000e+00 HA 14 1
The first number is the arbitrary number of protons given. The second and third numbers are the proton's chemical shift and the degree of uncertainty, if any. The fifth entry is the atom identifying the proton, and the sixth is the number of the residue present in the sequence to which this proton belongs. The fourth and the seventh entries are not used for our present purposes.
Assuming you have loaded your molecule, type turbo>nmr-spect. The program will then prompt you to give the proton file, the peaks file, and the reference peak file output names (with filename.peaks and filename.ref formats). The filename.peaks then contains a list of theoretical NOESY peaks, the relative intensities of which are deduced from the interproton distances determined on the loaded structure, assuming the existence of a rigid molecular structure.
This corresponds to a rough calculation of a theoretical spectrum. The peaks file generated can be directly used in EASY to display on a spectrum the location of theoretical NOESY peaks and thus to compare the theoretical and experimental locations. The main advantage of this option is that it helps end the assignment of NOESY spectra during the NMR refinement process.
The DGNL data base has the same structure as the Heap File. It contains the structural data on several molecules, with which a selected fragment of the loaded molecule can be compared. This data base already exists when you install Turbo-Frodo, but you can also create your own data bank by using the following protocol:
Generate a heap data called dgnl.heap by starting Turbo-Frodo. The program will ask you for the name of the Heap File. Type dgnl.heap. Then fill up this heap by using the following option, turbo>read pdb or turbo>readwh. You then have to stop the program. Restart it with another Heap File and type turbo>make-dgnl. This option calculates the distance matrix, and the output is the diag.mtx file. If you want to use the latest DGNL data base, you must copy the 2 files (dgnl.heap and diag.mtx) on the $TURBO_DIR/dict directory.