CoolFace
Datasetpublic

hugging-science/goa_reasoning

Dataset Description This dataset integrates Gene Ontology (GO) annotations (filtered to experimental evidence, IDA) with bibliographic data from Europe PMC.Each record links a biological entity (gene/protein) to a GO term and the supporting scientific publication (Title and Abstract). Columns DB, DB_Object_ID, DB_Object_Symbol — Source database and gene/protein identifiers Qualifier — Relationship qualifier (e.g., NOT, contributes_to) GO_ID, GO_Name — GO term… See the full description on the dataset page: https://huggingface.co/datasets/hugging-science/goa_reasoning.

sourceHugging Faceupdated 11mo agoView on Hugging Face
0likes24downloads
Dataset Card

Dataset Description

This dataset integrates Gene Ontology (GO) annotations (filtered to experimental evidence, IDA) with bibliographic data from Europe PMC. Each record links a biological entity (gene/protein) to a GO term and the supporting scientific publication (Title and Abstract).

Columns

  • DB, DB_Object_ID, DB_Object_Symbol — Source database and gene/protein identifiers
  • Qualifier — Relationship qualifier (e.g., NOT, contributes_to)
  • GO_ID, GO_Name — GO term and its descriptive name
  • DB_Reference — Supporting reference (PubMed or other)
  • Evidence_Code — Experimental evidence type (e.g., IDA)
  • Aspect — Ontology branch (P = process, F = function, C = component)
  • DB_Object_Name, DB_Object_Synonym, DB_Object_Type — Entity metadata
  • Taxon — Organism taxonomy ID
  • Date, Assigned_By, Annotation_Extension — Annotation metadata
  • PMID — PubMed ID for the supporting publication
  • Title, Abstract — Publication title and abstract text (fetched from Europe PMC)

Summary

This dataset forms a structured link between curated GO annotations and natural language descriptions of the underlying biology — suitable for:

  • Biological reasoning
  • Text mining
  • Literature-grounded machine learning (ML) datasets