Chemical Repeated-Dose Toxicity Data with OECD Test Guideline Validation

590.6 KB of data compiled from eChemPortal and J-CHECK databases for modeling subchronic oral repeated-dose toxicity. The dataset contains study-derived effect levels (NOAEL and LOAEL) for chemicals, with all underlying studies following OECD test guidelines 407, 408, and 422 under GLP conditions. Predictive models developed from this data achieved test set accuracies ranging from 0.642 to 0.682 and were experimentally validated by Global Product Compliance Europe AB.

Use Cases

• Train machine learning models to predict NOAEL values based on chemical structure data.
• Screen pesticides for potential repeated-dose toxicity using the Pesticides Properties DataBase (PPDB).
• Identify key toxic substructures, such as chlorine- and amine-group-containing aromatic systems, via substructure analysis.
• Conduct preliminary hazard screening for new synthetic organic chemicals as part of a weight-of-evidence evaluation.

Strengths

• All data originates from studies conducted under OECD test guidelines 407, 408, and 422 and Good Laboratory Practice (GLP) conditions.
• Models were experimentally validated using substances from Global Product Compliance (GPC) Europe AB.
• The final selected models for the test sets showed accuracies ranging from 0.642 to 0.682.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last updated 2026-03-25 18:03:44; freshness should be verified.

Provenance

Source

eChemPortal and J-CHECK databases.

Collection Method

Compiled for developing classification-based predictive models.

Freshness

Last updated 2026-03-25.