URSA-BBs: Commercially Available Building Blocks for Retrosynthesis

URSA Building Blocks (URSA-BBs) is a catalog of commercially available chemical building blocks used as stock-termination references in the URSA retrosynthesis benchmark. The dataset was created by insilicomedicine and was last updated on Hugging Face on July 7, 2026. It is described as a plain list of structures and identifiers without vendor or pricing metadata.

Use Cases

• Benchmarking retrosynthesis algorithms based on the URSA framework
• Validating predicted synthesis routes based on commercial availability of terminal compounds
• Identifying common chemical building blocks used in synthesis planning

Strengths

• Dataset is specifically designed for the URSA retrosynthesis benchmark
• Focuses on commercially available compounds, providing a practical constraint for synthesis planning

Limitations

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

Provenance

Source

insilicomedicine

Collection Method

Likely compiled from commercial chemical supplier catalogs.

Freshness

Last updated 2026-07-07 09:55:32; freshness should be verified