An integrated pharmacokinetics ontology and corpus for text mining

Area of Science:

• Pharmacokinetics and Pharmacogenomics
• Bioinformatics and Computational Biology
• Drug Discovery and Development

Background:

• Pharmacokinetic (PK) parameters, drug interaction data, and pharmacogenetics information are fragmented across databases and literature.
• Lack of a standardized pharmacokinetics ontology and annotated corpus hinders text mining for data collection and integration.

Purpose of the Study:

• To construct a comprehensive pharmacokinetics ontology to annotate diverse PK data.
• To develop a PK-corpus for text mining drug interaction and PK studies.
• To establish a hierarchical annotation scheme for key PK terms and interactions.

Main Methods:

• Developed a pharmacokinetics ontology covering in vitro/in vivo experiments and drug metabolism/transport enzymes.
• Created a PK-corpus with four classes of PK abstracts (in vivo/in vitro studies, pharmacogenetics, drug interactions).
• Implemented a three-level hierarchical annotation scheme for terms, sentences, and interaction pairs.

Main Results:

• The pharmacokinetics ontology successfully annotated in vitro and in vivo PK studies.
• The PK-corpus demonstrated utility in a drug interaction extraction text mining analysis.
• The annotation scheme effectively tagged key PK information and drug interactions.

Conclusions:

• The developed pharmacokinetics ontology provides comprehensive annotation for PK experiments.
• The PK-corpus serves as a valuable resource for text mining PK parameters and drug interactions.
• This work facilitates automated data extraction and integration in pharmacokinetics research.