Nice article, Angel. Too bad it's out of date already! :)_ No, really -- it is a nice article, and I'm delighted that the business I've been working on involving surfaces is so apropos. I think that should be the last nail in the coffin for Chime.
The business about two surfaces is really interesting and actually quite simple to implement in Jmol. Each surface is generated by an internally-derived voxel cube consisting of data points measuring the closest point of the voxel to the surface; 0 being at the surface, <0 being inside the surface; >0 being outside the surface. Now, mapping consists of using two such voxel cubes. The first defines the surface; the second defines the map. I hadn't thought about using the voxel data for the solvent business as a map, but when you consider it as a molecular surface, well, why not? The atoms selected for the mapping creates set of cube data that then determines the color of the surface data. What you will get is a coloration exactly as you describe -- red closest; blue furthest from the other surface. OK, so all this requires is the addition of a new keyword for the isosurface command: select (atom expression) This should override any previously selected set, and should be used twice: isosurface select (1-3) solvent map select (2-4) solvent (2 hours later) That's in now. It will take some experimentation to get this figured out. Lots of possibilities here. Bob Angel Herraez wrote: >Wonderful, Bob! > > > >>The idea comes from a conversation I had recently with Chris Larsen of >>Vecna Technologies. He was interested in proper depiction of selected >>binding sites in enzyme models. For these depictions one wants just the >>part of the overall molecular surface that corresponds to the binding >>site. And it could be useful to color different fragments different colors. >> >> > >I agree. This is how surfaces work (or at least can work, don't >remember now) in Chime and, among other things, opens the way for a >very interesting feature that Eric Martz designed into Protein >Explorer: the "contact surfaces", where a fragment of surface is >generated on either the ligand or the receptor, and the surface is >colored according to distance to the other moiety. For that, the >ability to color the surface in a stepwise gradient is needed (say, >select 5 A distance and color grey, then 3 A distance and color >magenta, then 2 A distance and color pink). > >Next week I may have time to specify this in more detail (or Eric >will do). > > > > >>Q: Do you like this idea? >> >> > >Definitely > > > >>Q: Does anyone want to retain the "old" way? >> >> > >Don't think so. They look nice, but they are false: the molecule >continues, but the surface closes. > > > >>Q: Should I allow for the option? >> >>Q: If so, which should be the default -- subselection gives fragment or >>subselection gives closed surface? >> >> > >Fragment. > > > > > > >------------------------------------------------------------------------- >Using Tomcat but need to do more? Need to support web services, security? >Get stuff done quickly with pre-integrated technology to make your job easier >Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo >http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 >_______________________________________________ >Jmol-users mailing list >[email protected] >https://lists.sourceforge.net/lists/listinfo/jmol-users > > ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Jmol-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jmol-users
