Pregnane-X Receptor Induction Activity for Over 11,000 Compounds

Over 11,000 chemical compounds were screened for their ability to induce the human Pregnane-X Receptor (PXR). The dataset was created by the OpenADMET project for a blind prediction challenge, with a challenge period starting in April 2024. It is the largest publicly available collection of PXR activity data.

Use Cases

• Predict PXR induction activity from molecular descriptors or fingerprints using regression models.
• Train binary classifiers to identify PXR-active compounds based on assay results.
• Benchmark machine learning models for ADMET property prediction against a standardized challenge dataset.
• Analyze structure-activity relationships for PXR ligands using chemical compound data.

Strengths

• Contains over 11,000 compounds, making it the largest public PXR activity dataset.
• Data was generated using a high-fidelity in-house assay, indicating experimental quality.

Limitations

• Specific column names, data types, and the exact number of features are not provided in the input.
• The dataset's temporal coverage and update frequency beyond the challenge period are unknown.

Provenance

Source

OpenADMET project (author: openadmet), hosted on Hugging Face.

Collection Method

Experimental screening of compounds using a high-fidelity in-house assay.

Time Range

null

Freshness

null

Geography

null