Begin forwarded message: From: Elaine Meng <meng@cgl.ucsf.edu> Date: January 8, 2008 10:09:06 AM PST To: Adam C Marko <adam.marko@ucsf.edu> Cc: Terry Lang <terry@lego.berkeley.edu>, sbrozell@scripps.edu, Eric Pettersen <pett@cgl.ucsf.edu>, Elaine Meng <meng@cgl.ucsf.edu> Subject: sphere viz in Chimera Hello, You guys will probably think I'm a nutjob (or prior suspicions may be confirmed), but Adam got me started thinking about how to improve sphere visualization in Chimera, and I wrote a short program akin to showsphere but that takes advantage of features in Chimera (attached at bottom - feel free to use, alter, redistribute, or ignore completely - I spot-checked the output but it's still caveat emptor). The old program showsphere was written in the time of Midas and shows the radii with a dot molecular surface. If you just show the PDB file with the sphere representation in Chimera, that doesn't include the radius information but just uses some default for carbon since the spheres are written as C atoms. The dot surface cannot be used for volume calculations in Chimera. Showsphere also made a separate file for each cluster. Chimera can show solid spheres, and an attribute assignment file can be used to set the radii from the SPHGEN output. Another difference from Midas is that you can read in multiple models from a single PDB file which are then separately displayable and specifiable. It was Adam's idea to take advantage of the controls in the Model Panel. So this new little program puts all the clusters (above a user-specified minimum size) into a single PDB file and creates an attribute assignment file for setting sphere radii in Chimera. The volume enclosed by a "molecular" surface around the cluster can be computed in Chimera. -- Example (using rec.sph, the SPHGEN output from the DOCK6 tutorial http://dock.compbio.ucsf.edu/DOCK_6/tutorials/sphere_generation/generating_s... ): f77 chimerasphere.f -o csphere guanine.24% csphere enter name of sphere file: rec.sph rec.sph minimum size to include? 35 35 enter name of output PDB: recspheres.pdb recspheres.pdb enter name of radius assignment file: recspheres.rad recspheres.rad cluster # 1 contains 158 spheres cluster # 2 contains 80 spheres cluster # 3 contains 50 spheres cluster # 4 contains 46 spheres cluster # 5 contains 38 spheres -- then I started Chimera and used the following commands before saving screenshot below: open rec_noH.pdb (also from DOCK6 tutorial) open recspheres.pdb defattr recspheres.rad (to browse to this file instead, use "defattr browse" or Tools... Structure Analysis... Define Attribute. This step generates message "No attributes usable by Render dialog were defined" but that is nothing to worry about; just means "radius" is one of the attributes suppressed from appearing there) rep sphere #1 rainbow models #1 color purple #0 set bg_color white -- screenshot with Model Panel (Favorites... Model Panel) -- note you can toggle display of clusters individually using the Model Panel I originally had the program ask for a model number into which the sphere PDB file would be read, because the attribute assignment file must specify each sphere. For example, if you specified 2, you would have to "open 2 recspheres.pdb" and the assignment file would include specifications like #2.5:1017.het@c The current version doesn't ask and just uses specs like #.5:1017.het@c for which there is a [remote] possibility of matching something other than the sphere, but nonetheless it could happen. Lines for the alternative behavior are still in the file, just commented out. Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html