Ginkgo Datapoints
    Releasing the GDPsm1 ADME Dataset
    Data

    Releasing the GDPsm1 ADME Dataset

    GDPsm1 profiles a 320-compound subset of the Library of Pharmacologically Active Compounds (LOPAC 1280) across key Tier 1 ADME assays

    Today, we’re excited to announce the release of GDPsm1.1, the first installment of our GDPsm1 ADME dataset. Download and access GDPsm1.1, along with additional multi-modal datasets, for free on the Ginkgo Datapoints portal.

    GDPsm1 adds to the reservoir of public domain ADME data published by companies like Biogen, whose seminal dataset has been invaluable as a resource in benchmarking ML models and advancing our collective understanding of small molecule developability. Ginkgo Datapoints addresses the need for ML fit-for-purpose, metadata-rich data for your ML-driven workflows. Our automation-first approach is built for the scale required to help you get answers and discover molecules faster.

    Dataset overview

    GDPsm1 profiles a 320-compound subset of the Library of Pharmacologically Active Compounds (LOPAC 1280) across key Tier 1 ADME assays, with GDPsm1.1 focusing on kinetic solubility.

    Each of the assays is run with technical replicates and robust QC to ensure data quality.

    Experimental design

    Each assay was executed in 384-well format and analyzed by acoustic ejection mass spectrometry (AEMS) on the SciEx Echo MS system coupled with a QQQ 6500 mass spectrometer in positive ionization mode.

    Kinetic solubility

    • Media: Phosphate buffer solution, pH 7.4
    • Incubation time: 2 hours
    • Equilibration temperature: 37°C\
    • Data readout: Solubility

    Microsomal stability

    • Species: Human (10-donor pooled) liver microsomes
    • Incubation time points: 0, 10, 20, 30, 45 minutes
    • Data readout: CLint, half life

    P450 inhibition

    • CYP isoform: CYP2D6
    • Test system: Recombinant enzyme
    • Probe substrate: Dextromethorphan
    • Test compound concentrations: 10, 3.33, 1.11, 0.37, 0.12, 0.04, 0.01 uM
    • Data readout: IC50

    More detailed methods are included within the dataset download.

    How to use GDPsm1

    Drug discovery & development

    • Use in vitro ADME data to predict downstream pharmacokinetics
    • Triage compounds for lead optimization based on profile of desirable or undesirable ADME characteristics (e.g., poor metabolism)

    Quantitative Structure-Activity Relationship (QSAR) modeling

    • Build predictive models by training models to predict ADME properties from molecular structures
    • Integrate with existing internal or publicly available ADME datasets to improve model generalization across a diverse chemical space
    • Combine with orthogonal datasets to create a holistic profile of drug candidate behavior

    How to access GDPsm1.1

    Researchers and developers can download the dataset directly from the Ginkgo Datapoints portal. For more information about our ADME data generation services, get in touch with the team.