CREED: Comprehensive Reactant Exhaustive Enumeration Dataset for Retrosynthesis

A large-scale, quality-controlled resource of chemical reactions for single-step retrosynthesis. It pairs target product structures with many plausible reactant sets, reflecting that synthesis planning is inherently multi-solution. The dataset was created by insilicomedicine and was last updated on Hugging Face in May 2026.

Use Cases

• Training machine learning models for retrosynthesis based on enumerated reactant sets.
• Evaluating the diversity of plausible synthetic pathways for a given product.
• Benchmarking single-step disconnection prediction algorithms against a multi-solution dataset.

Strengths

• Described as a large-scale resource, suggesting a substantial number of examples.
• Quality-controlled, implying a level of curation and validation.
• Explicitly designed to reflect the multi-solution nature of synthesis planning.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

insilicomedicine

Freshness

Last updated 2026-05-18 08:00:41; freshness should be verified.