OpenPHACTS: Identity Co-Reference Rules for Drug Discovery Datasets

A paper from the University of Manchester presents rules for computing co-reference chains across multiple drug discovery datasets. The work addresses the challenge of linking equivalent concepts across datasets that use different identifiers and data models. The approach emphasizes capturing the context of equivalence to allow data consumers to control which co-references are included in their applications.

Use Cases

• Resolving entity identity across disparate drug discovery datasets based on the described co-reference rules.
• Building context-aware data integration pipelines for pharmacology research.
• Evaluating different operational notions of equivalence for specific application domains in drug discovery.

Strengths

• Rules are derived from the OpenPHACTS Identity Management Service, a known project in the field.
• The work explicitly addresses the challenge of differing data models and user-specific notions of equivalence.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The license is closed, restricting reuse and distribution.

Provenance

Source

University of Manchester

Collection Method

Rules developed for the OpenPHACTS Identity Management Service.

Time Range

null

Freshness

Last update date is unknown; freshness unverified.

Geography

null