PREKIN is a utility program which prepares kinemages from Protein Data Bank format coordinate files. The kinemage can then be displayed, and modified if desired, in MAGE. PREKIN and MAGE are under continual development to increase their capabilities; use the latest versions, available free from the Web or ftp site at kinemage.biochem.duke.edu. Each version is identified in its name by number and platform: "PKIN_x_y.exe" for PC Windows, "PKIN_x_y.PPC" or "PKIN_x_y.68K" for the Mac, and (e.g.) PKIN_x_y.SGI on Unix, where "x.y" is the version number. This file is a guide to the process of using PREKIN to make your own kinemages. USING PREKIN version 5.4 (for MAGE version 5.4, released Mar. 1999): You need the PREKIN application, PDB-format coordinate files, a word processor for some editing of the *.kin files, and the MAGE display program to look at them. The easiest way to get PDB files is from the Web site at http://www.rcsb.org. The Make_kin.txt tutorial works from the PDB file 2aai; both are available from the kinemage Web site. The PREKIN interface uses a succession of dialog boxes to define the kinemage; there is a menu of built-in scripts in an early dialog box that allows easy selection of several useful kinemage layouts. PREKIN is aimed primarily at proteins and nucleic acids, but it also handles prosthetic groups, carbohydrates, waters, etc. Multiple passes through PREKIN can build a kinemage with more complicated groupings. The resulting kinemage is usually edited further to add, delete, recolor or rename things, either on-screen in MAGE or using a word-processor program. However, PREKIN output can always be immediately viewed and evaluated in MAGE. Both PREKIN and MAGE can handle large files, but to make an intelligible kinemage it is advisable to edit quite drastically at some stage of the process - for smooth rotation, but even more for comprehensibility. There are six types of control specifications in a PREKIN run, done through a series of dialog boxes. A given run may need only 1 or 2 of these, or could use all 6. 1) File selection for input and output is done immediately when PREKIN is started. 2) Choice of operation flow, as either a) the default "Backbone Browser" script (which will produce Calphas, SS, and non-water het groups for all subunits), b) a menu of built-in scripts, c) "New Ranges": specified sets of graphics objects and their ranges, d) "Focus": objects within a radius of a residue or a point, or e) a script read from an external file. In this dialog, one can also change some rarely-used options (e.g., correcting for left-justified atom names in the input file). For a fresh start if you make a mistake, the script-vs-range-vs-focus dialog box can be returned to from many of the following dialog boxes. 3) Built-in scripts presently include the following: a) all Calphas, plus disulfides and non-water het groups (same as Backbone Browsing); b) all mainchain, with Hbonds; c) sidechains by residue number; d) all sidechains, grouped by residue type; e) "lots": mainchain, sidechain, Calphas, any hydrogens, non-water hets, and waters (as balls), with B-factor in the point-IDs; and f) ribbon (simple version without explicit secondary structures). The scripts are designed either for use in "browsing" PDB files, or as starting points for an author. If in doubt, script a) is always a plausible choice to start with. For nucleic acids, script a) gives a virtual-bond pseudo-backbone drawn between P, c4', and c1' atoms; you can easily add a line to symbolize each base-pair, by using the drawline function in MAGE (suggest shortening lines by 1.3A) to connect the proper c1' atom pairs. Script d) on nucleic acids gives the pseudo-backbone, plus the bases, grouped and color-coded. Script e) gives all-atom backbone, sugars, and bases. H-bonds are not calculated between base pairs, but it is easy to add them in MAGE. For carbohydrates, which are treated as "hetatoms" in PDB files, PREKIN will look for all possible connections between sugar residues if you ask for that under "First: reset options" in the initial-choice dialog box. It can be slow, so use it only when needed. 4) Range controls: if you asked for ranges, you are given a range-control dialog box. For each range you specify the starting and ending residue numbers and/or the residue type ('res': e.g., lys) to set the extent of the range. Then you check the box for each kind of display object you want produced for that range (mc=mainchain, sc=sidechain, hb=backbone H-bonds, hy=hydrogens, ca=Calphas, ht= non-water heterogens, wa=waters, at=atom balls, lb=labels, ro=rotation or mutation; final set are ribbon controls). Then you click to accept that range, either going on to the next range, or ending the range set, or ending and writing a script file. (If Prekin doesn't find the hets or waters you expected, look in the coordinate file to make sure the beginning of the lines say "HETATM", rather than "ATOM ", and also check which subunit chainID they have). For example, to make a kinemage showing a beta hairpin you might enter the following two ranges: 14 to 35: mc, hb 24 to 25: sc That will produce vectors connecting mainchain N, Calpha, C, and O atoms, plus mainchain H-bonds for residues 14-35, and sidechains for residues 24-25. As another example, you could produce sidechain vectors and balls at the S atoms for just the Met residues in a protein, plus all Calphas, with the following ranges: -999 to 9999: met (replaces '...'), sc, at -999 to 9999: ca [You will get an additional dialog box asking whether you want balls or crosses to mark the atoms, and what radius for the balls (default is 0.2A).] To produce active-site sidechains for residues 34-37 and 86 of subunit A and residue 13 of subunit B, plus Calphas for both chains, you would need to do two passes (one for each subunit), since subunit choice applies to the entire set of ranges in a given pass. If you wanted to mutate residue 36 to an Arg and set up rotations for the sidechain chi angles, then you would enter the range: 36 to 36: arg (replaces '...'), ro (and hy if you wanted H atoms) [PREKIN will later ask for a library file to find an example of an Arg; you can open either a PDB file containing that residue, or else the standard PREKINAA.LIB file.] To make a ribbon with beta-strand arrows, helix spirals, and single-spline loops for file BPTI (like Kin. 4 of Demo5_4b.kin), enter the following ranges (chosen looking at a backbone kinemage, or from the PDB header): -999 to 9999: Ribbon, entire range 2 to 8: Ribbon, ALPHA range 47 to 57: Ribbon, ALPHA range 16 to 25: Ribbon, BETA range 27 to 36: Ribbon, BETA range 43 to 46: Ribbon, BETA range In the next dialog box, accept the defaults for ribbon width, etc. 5) Focus controls: used to make a display list of things within some radius of a specified point. Either the x, y, z of the focus point can be given, or PREKIN can find the center point of a given-numbered residue to serve as the focus point. PREKIN will then ask for radii within which it will output sidechains, mainchain, Calphas, waters, and non-water het groups; the default values for those radii are 8, 10, 15, 10, and 10 . The radius is tested at each atom of the target sidechain, mainchain, etc., which is then all included if any atom is inside. 6) Subunit selection: controls which subunits will be considered (recognized by the "chainID" field between aa name and residue number in the PDB file, or by a line starting with "MODEL n", as in multiple NMR structures). Each subunit will be put into a new group of display objects, with slightly different colors (which repeat after 5). The default for built-in scripts is sometimes just the first subunit in the file, but you can change that in this dialog box to specifiy either all subunits, or just the first, or those within a specified range. [Note: if you edit the subunit ranges, you must ALSO check the correct radio button.] For nucleic acids, request enough subunits to get both strands of a DNA duplex. After all the above items have been specified or defaulted, PREKIN will go to work and print messages about its progress, ending with a count of the number of triples written out and a message to select 'Open', 'New Pass', 'Launch kinemage', or 'Quit' from the file menu. A new pass allows you to do another run through script, menu, or focus selections, using the same input coordinate file and writing onto the end of the same output file; 'Open' lets you start a completely new kinemage. If you do not know residue numbers for the parts you want, then do an initial PREKIN run with the default Backbone Browser; then look at the resulting *.kin file in MAGE, pick atoms to identify the parts you want, and decide how to set up a second run of PREKIN. Even when the more complex scripts or ranges are used in PREKIN, the resulting kinemage generally functions as a starting version which you will want to edit and modify - saving views, changing colors, adding or deleting parts, and writing text and captions. For help with that process, Demo5_4b.kin is a brief introduction, Mage_5_4.txt documents the Mage on-screen editing capabilities, and Make_kin.txt is a more complete tutorial in the process of making and modifying your own kinemages. All of those, as well as up-to-date program versions, are available from the kinemage Web site at http://kinemage.biochem.duke.edu. The Unix version of PREKIN can be run from a command line. Type "prekin -help" for instructions.