Faculty, Staff and Student Publications
Language
English
Publication Date
1-25-2024
Journal
Studies in Health Technology and Informatics
DOI
10.3233/SHTI231043
PMID
38269887
PMCID
PMC13034061
PubMedCentral® Posted Date
3-31-2026
PubMedCentral® Full Text Version
Author MSS
Abstract
Automatic extraction of relations between drugs/chemicals and proteins from ever-growing biomedical literature is required to build up-to-date knowledge bases in biomedicine. To promote the development of automated methods, BioCreative-VII organized a shared task - the DrugProt track, to recognize drug-protein entity relations from PubMed abstracts. We participated in the shared task and leveraged deep learning-based transformer models pre-trained on biomedical data to build ensemble approaches to automatically extract drug-protein relation from biomedical literature. On the main corpora of 10,750 abstracts, our best system obtained an F1-score of 77.60% (ranked 4th among 30 participating teams), and on the large-scale corpus of 2.4M documents, our system achieved micro-averaged F1-score of 77.32% (ranked 2nd among 9 system submissions). This demonstrates the effectiveness of domain-specific transformer models and ensemble approaches for automatic relation extraction from biomedical literature.
Keywords
Electric Power Supplies, Knowledge Bases, Deep Learning, Drug-protein relation extraction, BERT, Ensemble Learning, Pubmed abstracts
Published Open-Access
yes
Recommended Citation
Das, Avisha; Li, Zhao; Wei, Qiang; et al., "Extracting Drug-Protein Relation from Literature Using Ensembles of Biomedical Transformers" (2024). Faculty, Staff and Student Publications. 905.
https://digitalcommons.library.tmc.edu/uthshis_docs/905