Mutagenicity Classification of Nitroaromatic Compounds

MUTAG is a dataset for binary classification of molecular mutagenicity. It contains nitroaromatic compounds labeled for their mutagenic effect on Salmonella typhimurium. The dataset was uploaded to Hugging Face by 'graphs-datasets' in February 2023.

Use Cases

• Predict mutagenicity labels from molecular graph structures using graph neural networks.
• Benchmark binary classification models using the provided 10-fold cross-validation protocol.
• Analyze the relationship between nitroaromatic compound structures and their toxicological endpoints.

Strengths

• Designed for a specific, well-defined binary classification task (mutagenicity).
• Includes a standardized evaluation protocol using 10-fold cross-validation.

Limitations

• Exact size (number of molecular graphs) is not specified in the provided information.
• Scope is limited to nitroaromatic compounds, which may not generalize to other chemical classes.

Provenance

Source

graphs-datasets on Hugging Face.

Collection Method

A collection of nitroaromatic compounds compiled for mutagenicity prediction.

Freshness

Dataset card was last updated in February 2023.