QSAR Fish Toxicity: 908 Chemicals with 6 Molecular Descriptors

QSAR_fish_toxicity contains 6 molecular descriptor attributes for 908 chemicals, used to predict acute aquatic toxicity for the fish Pimephales promelas (fathead minnow). The dataset was developed for quantitative regression QSAR models, using LC50 (the concentration lethal to 50% of test fish over 96 hours) as the target feature. It is licensed under CC-BY-4.0 and hosted on OpenML.

Use Cases

• Building regression models to predict LC50 toxicity based on molecular descriptors like CIC0 and MLOGP.
• Analyzing the relationship between 2D autocorrelations (GATS1i) and acute aquatic toxicity.
• Feature selection studies for QSAR models using atom-type counts (NdsCH, NdssC) and information indices.
• Benchmarking machine learning algorithms on a tabular dataset with a defined quantitative response (LC50).

Strengths

• Contains data for 908 distinct chemicals, providing a substantial sample size for model development.
• Includes 6 explicitly named molecular descriptors covering information indices, 2D matrices, autocorrelations, atom-type counts, and molecular properties.
• Has a clearly defined, quantitative target feature (LC50) measured in a standardized 96-hour test.

Limitations

• Row count and file format details are unknown, which may limit suitability assessment.
• Column-level documentation beyond the six descriptor names is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

OpenML platform

Collection Method

Data was used to develop quantitative regression QSAR models for predicting acute aquatic toxicity.

Time Range

null

Freshness

Last updated date is unknown.

Geography

null