BENT-PubMedBERT-NER-Gene

Named Entity Recognition (NER) model to recognize gene and protein entities.

Please cite our work:

@article{NILNKER2022,
  title = {NILINKER: Attention-based approach to NIL Entity Linking},
  journal = {Journal of Biomedical Informatics},
  volume = {132},
  pages = {104137},
  year = {2022},
  issn = {1532-0464},
  doi = {https://doi.org/10.1016/j.jbi.2022.104137},
  url = {https://www.sciencedirect.com/science/article/pii/S1532046422001526},
  author = {Pedro Ruas and Francisco M. Couto},
}

PubMedBERT fine-tuned on the following datasets:

• miRNA-Test-Corpus: entity type "Genes/Proteins"
• CellFinder: entity type "GeneProtein"
• CoMAGC: entity "Gene"
• CRAFT: entity type "PR"
• GREC Corpus: entity types "Gene", "Protein", "Protein_Complex", "Enzyme"
• JNLPBA: entity types "protein", "DNA", "RNA"
• PGxCorpus: entity type "Gene_or_protein"
• FSU_PRGE: entity types "protein", "protein_complex", "protein_familiy_or_group"
• BC2GM corpus- : entity type
• CHEMPROT: entity types "GENE-Y", "GENE-N"
• mTOR pathway event corpus: entity type "Protein"
• DNA Methylation
• BioNLP11ID: entity type "Gene/protein"
• BioNLP09
• BioNLP11EPI
• BioNLP13CG: entity type "gene_or_gene_product"
• BioNLP13GE: entity type "Protein"
• BioNLP13PC: entity type "Gene_or_gene_product"
• MLEE: entity type "Gene_or_gene_product"