FAIR data with KNIME

This workflow exemplifies how KNIME can be used to make data FAIR. The use case is a low-throughput screen conducted by academic partners. The results are published in this paper: https://doi.org/10.14573/altex.1712182.

The workflow that combines data and information from 50 individual data Excel files into one large data table. The CAS numbers as chemical identifiers were extended by SMILES, InChI and InChI keys using the RDKit KNIME community nodes and REST API to enhance interoperability. The metadata was extended by controlled vocabulary using REST API and programmatical access to chEMBL and chEBI databases. To comply with FAIRness, details about the used databases and ontologies are extracted as well. To give as much provenance as possible, user-defined metadata is added using the interactive Table Editor. Depending on the repository where the (meta)data should be deposited at, the workflow can be extended by automatic upload using a PUT request.