FAIRSCAPE

FAIRSCAPE implements provenance recording, Evidence Graph construction, persistent identifier assignment, and FAIR metadata annotation to enable transparent, reproducible computational research and generation of FAIR Evidence.

Key Features:

• Provenance Tracking: Records provenance of datasets, software, and computations including runtime parameters, environmental settings, and personnel contributions.
• Evidence Graphs: Constructs an Evidence Graph for each computational result linking persistent identifiers and rich metadata for software, computations, and datasets and storing a URI to the graph root in the result metadata.
• Ontology Support: Employs the EVI ontology (https://w3id.org/EVI) to support inferential reasoning over Evidence Graphs.
• Workflow Management: Manages nested and disjoint workflows while preserving provenance across Apache Spark jobs, scripts, and user-supplied containers.
• Persistent Identifiers: Assigns persistent identifiers to all objects, including software components, to enable unique long-term referencing.
• FAIR Metadata Annotation: Annotates results with FAIR-compliant metadata using the Evidence Graph model to enhance accessibility, validation, reproducibility, and reusability.

Scientific Applications:

• Large-scale computational analyses: Supports large-scale computational analyses where transparency and reproducibility are critical.
• Cross-disciplinary data processing: Documents and archives computational processes across diverse disciplines that rely on complex data processing workflows.

Methodology:

Records detailed metadata (runtime parameters, environment, personnel), constructs Evidence Graphs with persistent identifiers and a URI to the graph root, applies the EVI ontology for inferential reasoning, manages nested and disjoint workflows including Apache Spark jobs, scripts, and user-supplied containers, and annotates results with FAIR-compliant metadata.