Datafairy Bioassay Annotation

The Challenge

This project will:

• Revolutionize biotech R&D by standardizing research methods and improving reproducibility.
• Save time for scientists by reducing assay planning efforts and avoiding failed experiments.
• Help data scientists harmonize datasets, cleanse data, and enable advanced analytics, ML, and AI.
• Foster precompetitive collaborations and interoperable scientific data initiatives.
• Lower internal bioassay curation costs and simplify regulatory submissions.
• Aid assay kit vendors and CROs by increasing visibility in public databanks like ChEMBL and PubChem.
• Improve scientific publications’ quality through common assay annotation standards and public data banks.
• Maximize research funders’ ROI by increasing the value of funded science.

The project do this by addressing key issues:

• Over 1.4 million assay protocols in publications lack suitable formats for automated mining.
• Research organizations spend 4–12 weeks per assay selecting, setting up, and validating protocols, often leading to wasted efforts.
• Current manual curation methods are labor-intensive, while automated systems remain inaccurate.
• Many organizations already convert unstructured assay protocols into machine-readable forms, duplicating costly efforts.
• Obsolete assay protocols and evolving technologies hinder reproducibility and historical data interpretation.

What will the project deliver?

• Share costs for converting published bioassay data into accurate, machine-readable FAIR data using a community-defined model.
• Develop a FAIR data model based on public ontologies like BioAssay Ontology and promote it as an industry standard.
• Ensure FAIR data is publicly accessible after a brief exclusivity period for partners.