I really like the CDK 3D Viewer node, however it would be great if you could view multiple 3D structures in the window together.
For example this could be useful if you have already carried out some alignments of compounds onto each other such that their 3D coordinates are in the same area, and thus you may want to view these overlaid together by shift/ctrl clicking on multiple rows in the table.
Is this possible? If it is, it would be useful if 2nd/3rd/4th structures etc could have slightly different coloured carbon frameworks to easily distinguish the separate molecules.
good idea. I have enabled multiple selection and visualisation of molecules in the JMol viewer.
I'm not sure I can assign different colors to each molecule. For that I would have to dig into the JMol source code. Currently the common RasMol color scheme is applied to all molecules.
Also, there is a "maximum bond count": If you display many molecules, not all bonds are drawn for some reason. Haven't found a way yet to turn that off.
There is also the JMol Viewer in the Erl Wood Cheminformatics plug-in of course. Does the same thing really.
Hope that helps. Please let me know if the update works for you.
just to be sure, have you tried re-adding the same node and re-executing the workflow from scratch? I tried it myself and it started to work after that.
Probably to do with an configuration isse. I will look into it.
since we had the same problem showing aligned molecules we implemented a 3D viewer that can show molecules from two different input ports in parallel. Our BioSolveIT Viewer is part of our package, but can freely be use by everyone.
If you want to try it install the BioSolveIT KNIME node from:
Unfortunately I am still unable to get the 3D Viewer to display even after rerunning the workflow and reputting back in the 3D Viewer node. The table shows fine, but the 3D panel is just grey. I am trying to display molecules from an SDF column, on KNIME 2.9.1 on a Mac. It was working previously.
sorry to hear you had so much trouble with the node. I found the mistake: The autoconversion did not work properly. It should work if you use CDK molecules instead (e.g. via a Molecule to CDK node).
