negtox-llm-processed: Preprocessed NTD Drug Discovery Molecules

13,572 molecules across four diseases form this preprocessed dataset for drug discovery. It appears to be derived from the NTD (Neglected Tropical Diseases) dataset and has been processed using an LLM-based pipeline. The specific source, license, and update date are not provided.

Use Cases

• Train molecular property prediction models based on the described 13,572 molecules.
• Screen for potential drug candidates against the four mentioned diseases.
• Benchmark LLM-based preprocessing pipelines for chemical data.

Strengths

• Contains 13,572 molecules, providing a substantial base for analysis.
• Focuses on four specific diseases, offering a targeted scope.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known, but the specific features per molecule are unknown.
• Last update date is unknown; freshness unverified.

Provenance

Source

Likely derived from the NTD (Neglected Tropical Diseases) dataset.

Collection Method

Preprocessed using an LLM-based pipeline.