Miguel wrote: >> Hi all, >> >>I tried the "pmesh" functions of Jmol (10.00.36) and have a few questions. >> >>Currently there seems to be no labelling possibility for pmesh objects. >> >>Q: Would you consider adding a pmesh labelling command? >> > >It is not clear to me what kind of label you want. > >Q: Do you want a label that would show up on the display? > The same kind of label as for atoms, permanently shown (Are there other kinds? Or do you mean 'hover'?). >Q: How would one decide where it would be displayed? > The default reference point could be the first coordinate in the pmesh file. Optionally the number of the reference point coordinates in the pmesh file could be specified in the pmesh definition line (e.g.: pmesh example1 "example1.pmesh" 17). The rest would then be similar to atom labels. >Q: Could the labels be in one of the corners, in the same color as the >associated pmesh surface? > > If it would be in the corner in the same color it wouldn't be visible if the pmesh would be filled. And the label might also be too long to fit into the pmesh. >>In order to address pmesh objects individually it is necessary >>to put them into separate files. This is not very efficient >>if you have dozens of objects >>and need to contact the server for each object. >> >>Q: Would you consider adding support for a multiple pmesh format to be >>able to load all pmesh objects at once? >> > >Probably not. > >The pmesh format is a standard format. I do not want to create new file >formats. > >I don't think there is much inefficiency introduced by having to fetch >multiple files from the file server. If the .pmesh files are small then it >is not a big deal. If the .pmesh files are large then it really does not >matter. > >Web servers are quite efficient. Go to any commercial web site and dozens >of image files are downloaded, each of which is downloaded separately from >the web server. > > And then you will notice the difference if you switch off image download, depending on the usage and settings of the server. But the web server part is only one end of the story. Since we are providing information for all ~34000 PDB entries it would make a considerable difference to have additionally 34000 other files or 340000. But there are of course always several roads to rome...
>>The minor alternative to pmesh would be to add pseudo atoms to the >>coordinate >>file (which could be labelled) and use bonds to build up some kind of >>objects. >> > >I do not understand. > Maybe I should tell what we want to do. We want to highlight exon boundaries in protein structures. In order to be able to highlight for example PROSITE motifs at the same time the best way would be to show something between the two amino acids at the border. With pmesh this would be a filled square, perpendicular to the trace between the amino acids. With pseudo atoms in the corners of the square and bonds from each atom to each other atom you would get a square with a cross inside. Not as good as a filled square but at least it could be labelled and all squares could be integrated into the PDB file as additional model. > >>Q: Is it possible to change the appearance of bonds besides changing their >>color? >> > >You can adjust color, radius, and translucency. > >hbonds are displayed as segmented (dashed) cylinders > >Q: What other renderings/formats would be interesting? > > > When I tested the pseudo atom pmesh replacement I just didn't realized the possibility to adjust the bond radius by using the wireframe command (although I used it plenty of times before!). For their regular use I guess cylinders are sufficient. For their 'misuse' a flat rectangle would be nice, but then you would have the orientation problem. So just forget about it. Regards, Rolf ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ Jmol-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jmol-users
