Chemical entity normalization for successful translational development of Alzheimer’s disease and dementia therapeutics

• Published in: Journal of biomedical semantics Vol. 15; no. 1; pp. 13 - 12
• Main Authors: Mullin, Sarah, McDougal, Robert, Cheung, Kei-Hoi, Kilicoglu, Halil, Beck, Amanda, Zeiss, Caroline J.
• Format: Journal Article
• Language: English
• Published: London BioMed Central 31.07.2024; BioMed Central Ltd; Springer Nature B.V; BMC
• Subjects: Algorithms; Alzheimer; Alzheimer Disease - drug therapy; Alzheimer Disease - metabolism; Alzheimer's disease; Bioinformatics; Biological Ontologies; Care and treatment; ChEBI; Combinatorial Libraries; Computational Biology/Bioinformatics; Computer Appl. in Life Sciences; Data mining; Data Mining and Knowledge Discovery; Dementia; Dementia disorders; Dictionaries; Disease; Drug development; Entity normalization; Humans; International Society of Molecular Biology (ISMB) Bio-Ontologies Community of Special Interest; Mathematics; Mathematics and Statistics; Medical Subject Headings-MeSH; Methods; Names; Natural Language Processing; Neurodegenerative diseases; Ontology; String matching; Translational Research, Biomedical; United States; Entity normalization; Ontology; Alzheimer; ChEBI; Dementia
• ISSN: 2041-1480; 2041-1480
• DOI: 10.1186/s13326-024-00314-1

Background Identifying chemical mentions within the Alzheimer’s and dementia literature can provide a powerful tool to further therapeutic research. Leveraging the Chemical Entities of Biological Interest (ChEBI) ontology, which is rich in hierarchical and other relationship types, for entity normalization can provide an advantage for future downstream applications. We provide a reproducible hybrid approach that combines an ontology-enhanced PubMedBERT model for disambiguation with a dictionary-based method for candidate selection. Results There were 56,553 chemical mentions in the titles of 44,812 unique PubMed article abstracts. Based on our gold standard, our method of disambiguation improved entity normalization by 25.3 percentage points compared to using only the dictionary-based approach with fuzzy-string matching for disambiguation. For the CRAFT corpus, our method outperformed baselines (maximum 78.4%) with a 91.17% accuracy. For our Alzheimer’s and dementia cohort, we were able to add 47.1% more potential mappings between MeSH and ChEBI when compared to BioPortal. Conclusion Use of natural language models like PubMedBERT and resources such as ChEBI and PubChem provide a beneficial way to link entity mentions to ontology terms, while further supporting downstream tasks like filtering ChEBI mentions based on roles and assertions to find beneficial therapies for Alzheimer’s and dementia.