QSAR-TID-11: Quantitative Structure-Activity Relationships for Drug Toxicity

QSAR-TID-11 is a dataset curated by the TabArena team for evaluating predictive machine learning models on independent and identically distributed tabular data. The intended task is regression, and the data originates from a 2015 study on meta-learning for drug design. The original source is a Meta-QSAR study published by Olier et al. in 2018.

Use Cases

• Benchmarking regression models for independent and identically distributed tabular data based on the dataset's stated focus.
• Predicting molecular properties or toxicity endpoints based on QSAR descriptors implied by the platform tags.
• Meta-learning research in drug design based on the dataset's cited reference to the Meta-QSAR study.

Strengths

• Dataset is explicitly curated for evaluating ML models on IID tabular data, providing a clear benchmark purpose.
• Has a cited academic reference (Olier et al., 2018) and a traceable original data source on OpenML.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, sample data, and file formats are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Original data source: https://www.openml.org/search?type=data&sort=runs&status=active&id=3050

Collection Method

Curated from OpenML by the TabArena team for their Tabular ML IID Study.

Time Range

Dataset Year: 2015

Freshness

Dataset Year is 2015; last updated is unknown.

Geography

null