EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A large-scale evaluation of NLP-derived chemical-gene/protein relationships from the scientific literature: Implications for knowledge graph construction

Jonathan C G Jeynes iD, Matthew Corney, Tim James

PLOS ONE · 2023

DOI: 10.1371/journal.pone.0291142 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, 2023. Verfasst von Jonathan C G Jeynes und 2 weiteren Autor:innen. Beteiligte Einrichtung: Evotec (UK) Ltd., in silico Research and Development, Milton Park, Abingdon, Oxfordshire, United Kingdom. Hauptthemen laut MEDLINE: Natural Language Processing, Pattern Recognition, Automated. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

One area of active research is the use of natural language processing (NLP) to mine biomedical texts for sets of triples (subject-predicate-object) for knowledge graph (KG) construction. While statistical methods to mine co-occurrences of entities within sentences are relatively robust, accurate relationship extraction is more challenging. Herein, we evaluate the Global Network of Biomedical Relationships (GNBR), a dataset that uses distributional semantics to model relationships between biomedical entities. The focus of our paper is an evaluation of a subset of the GNBR data; the relationships between chemicals and genes/proteins. We use Evotec’s structured ‘Nexus’ database of >2.76M chemical-protein interactions as a ground truth to compare with GNBRs relationships and find a micro-averaged precision-recall area under the curve (AUC) of 0.50 and a micro-averaged receiver operating characteristic (ROC) curve AUC of 0.71 across the relationship classes ‘inhibits’, ‘binding’, ‘agonism’ and ‘antagonism’, when a comparison is made on a sentence-by-sentence basis. We conclude that, even though these micro-average scores are modest, using a high threshold on certain relationship classes like ‘inhibits’ could yield high fidelity triples that are not reported in structured datasets. We discuss how different methods of processing GNBR data, and the factuality of triples could affect the accuracy of NLP data incorporated into knowledge graphs. We provide a GNBR-Nexus(ChEMBL-subset) merged datafile that contains over 20,000 sentences where a protein/gene-chemical co-occur and includes both the GNBR relationship scores as well as the ChEMBL (manually curated) relationships (e.g., ‘agonist’, ‘inhibitor’) —this can be accessed at https://doi.org/10.5281/zenodo.8136752 . We envisage this being used to aid curation efforts by the drug discovery community.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Natural Language ProcessingPattern Recognition, AutomatedArea Under CurveDatabases, FactualDrug Discovery

Autor:innen und Institutionen

  • Evotec (UK) Ltd., in silico Research and Development, Milton Park, Abingdon, Oxfordshire, United Kingdom.

ORCID-Profile: Jonathan C G Jeynes

Zitieren

APA

Jeynes, J. C. G., Corney, M., & James, T. (2023). A large-scale evaluation of NLP-derived chemical-gene/protein relationships from the scientific literature: Implications for knowledge graph construction. PLOS ONE. https://doi.org/10.1371/journal.pone.0291142

Vancouver

Jeynes JCG, Corney M, James T. A large-scale evaluation of NLP-derived chemical-gene/protein relationships from the scientific literature: Implications for knowledge graph construction. PLOS ONE. 2023. doi:10.1371/journal.pone.0291142

BibTeX
@article{Jeynes2023,
  author = {Jeynes, Jonathan C G and Corney, Matthew and James, Tim},
  title = {A large-scale evaluation of NLP-derived chemical-gene/protein relationships from the scientific literature: Implications for knowledge graph construction},
  journal = {PLOS ONE},
  year = {2023},
  doi = {10.1371/journal.pone.0291142}
}
Kostenlos starten →