high-energy conformers and perspectives

I recently tried to use MarvinSketch to create ethane in its eclipsed conformation in a sawhorse perspective and with 3D information. I couldn't use the Select conformer tool, because the eclipsed conformer is a maximum. Eventually I solved the problem by making bicyclo[2.2.2]octane, going into Select conformers, choosing the desired perspective for a CH₂-CH₂ group, importing the view back into MarvinSketch, deleting most of the molecule, and changing the bridgehead C's into H's. NOT a very efficient way to do it!





Will there soon be a way to set dihedral angles about certain bonds or to start and stop rotation about certain bonds, so that conformers that are not minima can be generated? I should also be able to use the 3D viewer to choose a certain perspective for a structure, then have that perspective transfer back into MarvinSketch with the corresponding coordinates. (I asked for this feature a while back, but it does not appear yet to have been implemented.)

Bob,


Don't tell anyone because this is top secret information ;-)


We are going to build a new tool (probably called Marvin3D). It will be a 3D viewer with some editing capabilities (this will not be an editor) for small and large molecules. It will support the features you propose and many others, like distance and angle measurement, molecular surfaces, sophisticated shading, selection and highlighting individual objects, ribbon diagrams, etc. We expect the first release by next spring. This tool will probably be based on Java3D, in wich case the Java3D package will be needed to be installed on the client machines.


Do you think Marvin3D will be useful for you?





We are also considering to add the rotation feature into MarvinSketch. This task (and of course tons of others) was added to our internal developer forum when you sent your request.

Ferenc wrote:

Bob, Don't tell anyone because this is top secret information ;-) ... Do you think Marvin3D will be useful for you?

My lips are sealed.





Yes, it would be very useful to me. I look forward to seeing it!





Are you aware of the open-source Java program for 3D display of molecules, Jmol? You might save some effort and focus on more advanced features if you are able to use some of Jmol's code. I don't know anything about open-source rules, so it may not be possible, but I thought I'd raise the possibility.





My interest in Jmol is to convert it into a MacOS X browser plug-in, like Chime is a plug-in for Netscape 4.7. But so far I'm unable to interest anyone in doing it. :-( So the Mac currently has no browser plug-in (as opposed to an applet delivered by a server) capable of displaying 3D chemical structures. :-( I don't suppose you'd be interested in making a browser plug-in version of Marvin3D ...?





Regards,





Bob





P.S. Speaking of requests, I thought you should know that the lack of lone pair display in Marvin is confusing some of my students. Some of them discovered that if they set an atom to Radical -> Divalent, they can get a pair of electrons to appear next to an atom. They don't realize that they've turned the atom into a carbene (nitrene, oxene), because they don't yet know about such species. This problem is another argument for adding an automatic lone pair feature (with an on/off switch, like implicit H atoms) and for a total-electron-count or unshared-electron-count feature that interacts with the Radical flag.

bobgr wrote:

Are you aware of the open-source Java program for 3D display of molecules, Jmol? You might save some effort and focus on more advanced features if you are able to use some of Jmol's code. I don't know anything about open-source rules, so it may not be possible, but I thought I'd raise the possibility.

Jmol is a nice tool, but we need to support features that probably would not be possible or very difficult to add to JMol.

bobgr wrote:

My interest in Jmol is to convert it into a MacOS X browser plug-in, like Chime is a plug-in for Netscape 4.7. But so far I'm unable to interest anyone in doing it. :-( So the Mac currently has no browser plug-in (as opposed to an applet delivered by a server) capable of displaying 3D chemical structures. :-( I don't suppose you'd be interested in making a browser plug-in version of Marvin3D ...?

What does a browser pugin do that an applet cannot do?

bobgr wrote:

P.S. Speaking of requests, I thought you should know that the lack of lone pair display in Marvin is confusing some of my students. Some of them discovered that if they set an atom to Radical -> Divalent, they can get a pair of electrons to appear next to an atom. They don't realize that they've turned the atom into a carbene (nitrene, oxene), because they don't yet know about such species. This problem is another argument for adding an automatic lone pair feature (with an on/off switch, like implicit H atoms) and for a total-electron-count or unshared-electron-count feature that interacts with the Radical flag.

Presently the only feature related to this issue is Edit >> Add lone pairs, which will be available in Marvin 3.5. You can test it here


http://www.chemaxon.hu/shared/alpha/marvin/doc/dev/example-sketch1.1.html


Is this unusable for you?

Ferenc wrote:

What does a browser plugin do that an applet cannot do?

An applet must be delivered by the server to the local user. Currently, if I browse to a chemical structure file such as a .mol file, and the server (e.g., the American Chemical Society journals' Web sites) does not deliver an applet such as Marvin or Jmol to me, I have to copy the file's URL and paste it into Jmol running independently, or I have to download the file to my computer and then open it with Marvin, Jmol, etc. By contrast, a browser plugin is present on the local user's computer. I tell my browser to use the plugin whenever I browse to files with certain extensions or MIME types, such as .mol. When I browse to such a file, the plugin automatically launches and displays the structure in the browser window. I would like to view .mol files and the like with a plugin within Safari as smoothly as I do .pdf files (with PDF Browser) and as smoothly as I used to do .mol files (with Chime) in Netscape 4.7.





Safari apparently follows the conventions laid out by Netscape for plugins. See http://developer.apple.com/technotes/tn/tn2020.html and http://developer.netscape.com/docs/manuals/communicator/plugin/ . Also see http://www.codecomments.com/archive244-2004-4-153001.html .

Ferenc wrote:

Presently the only feature related to this issue is Edit >> Add lone pairs, which will be available in Marvin 3.5. You can test it here http://www.chemaxon.hu/shared/alpha/marvin/doc/dev/example-sketch1.1.html Is this unusable for you?

Unfortunately, yes. The label "LP" and the bond thereto makes it look like it's a chemical group, not a lone pair. At the very least could you just show a couple of dots instead of the LP? Even then, the bond has got to go. Treating the lone pair like an atom just won't meet my needs.

1) Plugin issue:





In Windows you don't even need a plugin, just installed Marvin Beans. If you click a URL link that points to a file with an extension associated to MarvinView, then Marvin View will be popped up with the structure.





Mac is trickier. We have not solved extension association yet, but even that wouldn't help much in the case of Safari. At the moment the best you can do is to download the file and open it with the MarvinView application (in Marvin Beans). We will examine this issue.





2) Lone pair issue


We are aware of the issue, but we don't have capacity to solve this in the very near future.

Hi, I have an issue when I try to generate conformers of somo aromatic radical molecules (thiophenyl radical-like). When I ask the program for conformers generation it appears a pop up window and the calculus gets down. I wonder if this happens because the force field (MMFF) is not allowed for calculations with open shield systems, or what other things could be happen?

Dear Estefania, 

we could not reproduce the issue of conformer generation. Can you help us with an example molecule?

Thank you, 

Daniel