On django: pymol, coot, ccp4mg, bobscript, raster3d,
On claude/gilmour:(SGI Unix) pymol, coot, insightII, molscript, bobscript, grasp, raster3d, ghostscript/ghostview.
On PCs: Adobe Illustrator, Adobe Acrobat Adobe Photoshop, Microsoft Powerpoint, coot, rasmol, VMD, Cn3D, ghostscript/ghostview, ChemDraw, pymol
Don't make a postscript file until you must. It's a dead end. You will lose clarity when you translate it to another format. If you must transfer from postscript, use ghostview (NT version is easiest to use). Input your postscript file, then choose print. Choose one of the tiff options as a device. This will convert to a tif file. The proper choice depends upon whether your figure is color or black/white. Choose print to file, remember to choose a reasonable dpi setting. You can also write jpeg and png files this way. After the tiff file is written it can be imported into photoshop, illustrator or powerpoint or used as is if that's what the journal wants.
Use the interactive version of molscript (molscript -gl) on claude.. Rotate the molecule until you're happy with the view, right click and print orientation. Copy this orientation into your molscript/bobscript iinput file. You may need to change the center or scaling for the optimum view.
coming
tiff files have become proprietary, so the linux libraries can't write them. render will now write png files (as pymol does.)
Two ways: (1) Edit the r3d file before running render to change the background from black to white (change the line 1 1 1 to 0 0 0).
(2) If you have a file called header.r3d in the directory you run bobscript from, those specifications will be placed into the raster3d input file.
Molscript requires an END at the end of each pdb file read. Edit your pdb file to add this line and all will be OK.
The program to use is rastep (part of raster3d). An arbitrary view is easy. A specific view can get a bit involved. Here are some hints.
Make a pdb file that contains only those atoms you wish to have in the figure. Run rastep as in the raster3d documentation (see help page). Render the scene using render in raster3d. If it's the view you want, you're done. If not, go on to the next paragraph.
You must determine the view you want using molscript -gl, then write out the rotation/transformation matrix and put it into the header of the input file for raster3d. Put the TRANSPOSE of the matrix that molscript gives you into the raster3d header file. At this point, you may get a blank output tiff file. Don't panic yet. Look at the render output. It will give you a new translation. Put these numbers into the 4th line of the matrix. Now your view should be OK.
If you just want to look at the ellipsoids, use coot.
Use pymol. See the hints.
Change the scale. Edit the raster3d file. The last number in the 4th line of the view matrix contains this factor. Make this a smaller numer to make the contents fill the box.
In molscript, label as you normally do, and use the stereo3d procedure of raster3d. The labels will be "floating" right next to the atoms as long as the z offset is 0.0.
In pymol, see hints.
Go to the journal web page and see what the requirements are. There will be restrictions on dots-per-inch, color space, size of the prints, etc. When you prepare your figures with molscript, bobscript, photoshop, whatever, make sure the original prints have AT LEAST as much resolution as the journal requires.
At this point, you need photoshop. Photoshop runs on kodaly. Choose CMYK color space. When you open your figure, it will ask you if you want the color space converted. Say yes. Your colors may change a bit here. If the figures aren't true color (photographs), it probaby doesn't matter. Resize the figure WITHOUT resampling. Do not resample until you are at the final figure size. When you get the figure as you want it, then resample at 300 dpi. (I hope the figure you are resampling has at least 300 dpi). Write a tif file or whatever type of file the journal wants and you're done.
Coming