RDKit From Molecule

Generates RDKit molecule column from a molecule string representation (SMILES, SDF or SMARTS) and appends it to the table. Depending on the input format of the molecule the usage of some options is not possible. All grayed out options are not taken into account when the RDKit molecule gets generated, regardless if flagged or not.

Options

Molecule column

The column that contains the molecules (SMILES, SDF or SMARTS type column). If the input column type supports multiple molecule types, the node will take SMILES before SDF before SMARTS.

Treat as query

Flag that can be used for SMILES and SDF input columns. Select it to prepare the resulting RDKit molecules as queries. This forces a full sanitization, keeps hydrogens in the molecule and merges query hydrogens.

New column name

Name of the new column in the output table.

Remove source column

Toggles removal of the input molecule column in the output table.

Error Handling

If the molecule in an input row cannot be converted, you can select if this row should be separated from the output table and be delivered at the second port, or if a missing value should be inserted instead of the RDKit molecule in the result table.

Also, it is possible to get additional error information added for all structures that failed conversion. If this option is checked, you need to specify the name of the error column.

Generate 2D Coordinates

Select this option to generate 2D coordinates for the molecules. The coordinates are used for, among other things, 2D renderings of the structures. If not selected, the renderer will (re-)compute coordinates on demand.

Force Generation

Select this box to enforce 2D coordinate generation. Coordinates may already be available in the (SDF) input, so this option allows you to discard the original coordinates and recompute from scratch.

Advanced

Keep Hydrogens

Switch this option on to prevent the RDKit from removing hydrogens from molecules constructed from SDF. This option has no effect for SMILES or SMARTS input.

Partial Sanitization

The RDKit does a fair amount of work when compounds are loaded to make sure that they are chemically reasonable and correct. This is called sanitization. When switching on partial sanitization the following options allow specific aspects of the sanitization to be turned off:

Strict Parsing of Mol Blocks

When SDF molecules are converted into RDKit molecules, this option sets the tolerance level for parsing mol blocks. The default value for new nodes is defined in RDKit Nodes / RDKit Types preferences. (Introduced in December 2022)

Partial Sanitization Option - Reperceive Aromaticity

If enabled (the default) all molecules are converted to a Kekule form and then Hueckel's rules are applied to determine aromaticity. If disabled, whatever aromaticity information is present in the input will be used. NOTE: SDF/MOL data contains no information about aromaticity.

Partial Sanitization Option - Correct Stereochemistry

If enabled (the default) all stereochemistry specifications will be checked to ensure that they are not redundant or that stereochemistry markers are not set on either atoms or double bonds that should not have them.

Views

This node has no views

Workflows

• _01_ChemicalLibraryEnumerationKNIME Hub
• (3D) Fragment library design workflowKNIME Hub
• 00_SetupKNIME Hub
• 00_SetupKNIME Hub
• 00_SetupKNIME Hub
• Show all 336 workflows

Links

• Bioisosteric replacement using SMARTS (KNIME and RDKit)by Antoine_Lacour on 2020-10-13

  forum.knime.com/p/105493

Developers