Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.

## jmol-users

 [Jmol-users] pdb and xyz From: hens borkent - 2004-02-26 16:39:56 ```Miguel, > Hens, what computation package are you using to generate your files? > Most of the time I use MOPAC and its options to scan a coordinate (distance, dihedral angle). Then I read the output into MOLDEN (not only because I share the office with its author) that will write out an m-XYZ file. With Gaussian the same path is followed. (B.t.w. we also have a MOPAC web interface that will do the same directly, all perl that extracts the structures and writes the m-xyz format, creates a web page with the animation, control buttons etc.) MOLDEN does write the m-XYZ for vibrations as well. About the bonding information in an animation: the point is that it can change from frame to frame. Think of the Diels-Alder reaction, where bonds move from none to single, single to double and vice versa. That's what I had in mind when I wrote 'editing by hand'. Complicated but not impossible to be handled by a script, provided you feed it with accurate distance criteria. Although you (Miguel) wrote that JMol is not a bonding tool: Personally I'd like see the gradual change, say from single to double, by splitting the bond, or playing with the transparency. We tried this in Java3D and it worked. Hens ```
 [Jmol-users] Re: MODEL term appropriate? From: hens borkent - 2004-02-26 17:34:57 ```:tim wrote: > I would like to disucss the use of the term MODEL in the context of > animations. First of all let's agree on "animation" in this context. The visualization of a sequence of structures, normally slightly different ones, that can be run forward, backward, continuous or step by step? Each structure is called a "frame" in Chime. Frames can be selected by their sequential number too. Up to now I haven't seen a fundamental border between a pdb-model and a xyz-frame, or any other format in which separate structures can be distinguished. It is Chime that draws a line between them. In Chime frames can not be shown on top of each other. However if that can be controlled in JMol by a script command "set models nmr|anim" ( set frames nmr|anim ?) it would be a good solution to distinguish between the two applications. Chime requires that all framed structures in a file have the same atoms in the same order. Not so for pdb-models I guess, I'm not sure though. Is this a restriction we would like to question? Is it present in JMol? Hens ```
 Re: [Jmol-users] Re: MODEL term appropriate? From: Miguel Howard - 2004-02-26 18:08:28 ```>> I would like to disucss the use of the term MODEL in the context of >> animations. > > First of all let's agree on "animation" in this context. Very good. The terminology will be very important in these discussions. > The visualization of a sequence of structures, normally slightly > different ones, that can be run forward, backward, continuous or step by > step? > Each structure is called a "frame" in Chime. > Frames can be selected by their sequential number too. And the 'animation' runs through the 'animation on' command ... it does not require any additional script commands. > Up to now I haven't seen a fundamental border between a pdb-model and a > xyz-frame, or any other format in which separate structures can be > distinguished. I agree 100%. So we are trying to eliminate that distinction in Jmol. > It is Chime that draws a line between them. > In Chime frames can not be shown on top of each other. > However if that can be controlled in JMol by a script command "set > models nmr|anim" ( set frames nmr|anim ?) it would be a good solution to > distinguish between the two applications. Why is there a need to distinguish between the two? There may even be some value in allowing someone to 'animate' between several models in an nmrpdb file as a simple type of 'morphing'. > Chime requires that all framed structures in a file have the same atoms > in the same order. > Not so for pdb-models I guess, I'm not sure though. Probably correct. > Is this a restriction we would like to question? Is it present in JMol? My plan is to try to do something reasonable ... not crash. If it looks bad then don't play it as an animation. Please let me know if you see any fundamental problems with this. Miguel ```
 Re: [Jmol-users] pdb and xyz From: Miguel Howard - 2004-02-26 17:09:42 ```Hens wrote: > Most of the time I use MOPAC and its options to scan a coordinate > (distance, dihedral angle). > Then I read the output into MOLDEN (not only because I share the office > with its author) that will write out an m-XYZ file. With Gaussian the > same path is followed. > > (B.t.w. we also have a MOPAC web interface that will do the same > directly, all perl that extracts the structures and writes the m-xyz > format, creates a web page with the animation, control buttons etc.) > > MOLDEN does write the m-XYZ for vibrations as well. I am not familiar with any of these tools, but it is helpful for me to have a little understanding of the process. > About the bonding information in an animation: the point is that it can > change from frame to frame. > Think of the Diels-Alder reaction, where bonds move from none to single, > single to double and vice versa. That's what I had in mind when I wrote > 'editing by hand'. Correct. And I think that it is important that this bonding be controlled by the author, or by a better tool than Jmol. > Complicated but not impossible to be handled by a script, provided you > feed it with accurate distance criteria. > Although you (Miguel) wrote that JMol is not a bonding tool: > Personally I'd like see the gradual change, say from single to double, > by splitting the bond, or playing with the transparency. We tried this > in Java3D and it worked. I agree with this 100%. Transparency is currently very difficult for me to work with ... the current graphics engine does not support it. I have been thinking about the frame interpolation when a bond disappears. That is, when the bond is present in one frame but not in another. I was thinking of have the size of the bond reduce to zero during the interpolated frames. I had not thought about the double-to-single bond transition, but I trust that we can come up with something. Miguel > Hens ```