BioSample dataset for ontology mapping evaluation

Datasets for the evaluation of automatic curation of BioSample using Large Language Models (LLMs).

Files

• biosample_cellosaurus_mapping_gold_standard.tsv
  The gold-standard dataset for the cell line ontology mapping task. This was manually created by the authors.
  
  
• biosample_cellosaurus_mapping_testset.json
  The BioSample dataset used for the cell line ontology mapping task. This was collected using the EBI BioSamples API.
  
  
• biosample_cellosaurus_mapping_result_llm_assisted.tsv
  The result of the cell line ontology mapping of the test dataset by the LLM-assisted pipeline.
  
  
• biosample_cellosaurus_mapping_result_metasra.tsv
  The result of the cell line ontology mapping of the test dataset by directly using the MetaSRA pipeline.
  
  
• biosample_gene_extraction_testset.json
  The BioSample dataset used for the gene name extraction task. This was collected using the EBI BioSamples API.
  
  
• biosample_gene_extraction_result.tsv 
  The result of the gene name extraction from the test dataset by the LLM-assisted pipeline.

Licenses

biosample_cellosaurus_mapping_gold_standard.tsv is licensed under CC BY 4.0.

All other files are licensed under CC0.