Selected Atoms get lost

When taking an SDF, it is possible to highlight one or more atoms within it for manipulation at a future point. (From in-house nodes and MOE nodes (Atom Selector)).
However I notice that if this SDF with selected atoms is converted to Indigo and then back to SDF, the settings for these selected atoms become lost.
Is it possible to have this feature retained through the Indigo nodes. I assume this information is kept in the SDF block, which gets lost during the Indigo conversions.

Hi Simon,

you're right. The selected atoms are stored in the sdf block as:

>  <selected_atoms>


where 1 2 and 3 are the indices of the selected atoms.

I hope this information helps others to support that feature.

Best regards


Hello Simon and Guido,

I added such functionality to the Indigo nodes 1.1.12. You can update them from the nightly builds.
I added it in the following manner:
1. There is a new Highlighter node that can highlight atoms and bonds according to column with atoms and bonds indices. There indices can be specified as a collection, or as a string with values separated by space/comma. Also this node has an option to clear highlighting.
2. There are new properties in the Properties node: "Highlighted atoms" and "Highlighted bonds". They return a string with highlighted indices. Should these properties return a collection of values instead of simple string?
Does this functionality works for you?

Hi Mikhail,

Thanks for the implementation of this, its working great. I see this works well with the substructure highlighting that was already implemented a while back, which is super.

For the output as string from the Molecule Property node, I think this is fine. I can see positives and negatives for either string or collection column. As the output is comma separated it could easily be transformed to a collection column later if needed.

I notice the highlighted atoms are still present after converting out from Indigo to SDF which is just what I wanted.

Thanks for this,


Hi Simon,

Yes, Molfile format supports highlighting and we are saving highlighted atoms and bonds into molfile. One potential issue is that if a molecule is saving in SMILES then atom ordering can be different and the "highlighted atoms" indices can be invalid. But we are saving highlighted atoms in SMILES too via extension string.