ORD Data: Open Reaction Database in Protobuf Format

The Open Reaction Database (ORD) contains chemical reaction data stored in Google's Protobuf binary format for efficiency. The data is stored in compressed *.pb.gz files but can be converted to a human-readable text format. The dataset was last updated on 2026-05-12.

Use Cases

• Train machine learning models for reaction outcome prediction based on structured reaction data.
• Analyze chemical reaction patterns and trends using the standardized ORD schema.
• Develop data processing pipelines for converting binary Protobuf reaction records into other formats.

Strengths

• Data is stored in an efficient, compressed Protobuf binary format (*.pb.gz).
• The dataset is part of the Open Reaction Database, a known scientific resource.
• Data can be converted to a human-readable text format (*.pb.txt).

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

open-reaction-database

Freshness

Last updated 2026-05-12 22:36:10; freshness should be verified.