Update to v1.36.2 - Bugfixes, updates to RDKit, minor enhancements

We have just released v1.36.2 of our community contribution on the nightly build channel. The release will propagate through to the stable build channels for KNIME 4.6 and KNIME 4.7 in the coming days.

Java Changes

This release updates the build to use Java 17 to be compatible with the latest RDKit release (which is now the minimum required version of RDKit - v4.6.1)

RDKit Changes

The RDKit update introduces some minor changes to node output:

  • SVG rendering looks essentially identical, but the SVG may differ slightly (and seems to differ between different platforms)

  • SMILES canonicalisation has changed - if you have MMP fragmentations stored you will need to regenerate these

  • Morgan fingerprints show slight differences - again, this will require an update to stored MMP fragmentations. It only appears to apply to fragmentations with explicit hydrogens (either because your incoming molecules had them and you didnt opt to remove them, or because you opted to add them and not remove them, or for and `[*]-[H]` fragmentations which require an explicit hydrogen

We also believe this update fixes the long-standing annoyance described at

Hi, i’ve istalled the vernalis plugin but it seems not compare in Node Repository. On the start knime 4.3.0, I receive this error message: ERROR RepositoryManager Node ‘com.vernalis.knime.mmp.nodes.loop.CutTypeLoopStartNodeFactory’ from plugin ‘com.vernalis.knime.chem.mmp’ could not be created. The corresponding plugin bundle could not be activated! ERROR RepositoryManager Node ‘com.vernalis.knime.mmp.nodes.uniquifyids.UniquifyIdsNodeFactory’ from plugin ‘com.vern…

Changes to the Collection handling nodes

Following a bug report from @mwiegand described at

Good morning, I happen to recognize a few node inconsistencies amongst which also surfaced. Whilst many nodes can be configured and execute even if not corresponding column type is present, the Collection to String can be configured but fails to execute. [Bildschirm­foto 2023-01-08 um 10.49.41] Suggestion In order to ensure workflows can proceed processing (soft fail), as data might not be consistently present or compliant, I’d like to suggest to add an option make make it default behavio…

all our collection handling nodes which accept multiple collection input columns via a column filter dialog panel now have a new option to ‘Fail if no columns selected’ - the legacy behavior is preserved where no value for this setting is found. One small consequence is that when this option is set to 

true

, if there are no collection columns in the incoming table then the node will now give a different error message to previously

We have also taken the opportunity to clean-up the layout of some of the node dialogs in this set - if you use them regularly, click carefully - some options have change how they are laid out slightly!

We also discovered a bug in which the 

List to Set

node could add an extra value to the output Set in an unlikely combination of circumstances (those being that the incoming list was represented internally as a Sparse List and the default value for the list was not actually present in the list) - this has also been fixed.

There is a full Changelog at

github.com

If you spot anything untoward please do let us know

Thanks

Steve