Authors: María Herrero-Zazo, Isabel Segura-Bedmar, Paloma Martínez
6th International Biocuration Conference (April 7-10, 2013, Cambridge, UK)
An Ontology for formal representation of Drug-Drug Interaction Knowledge
Delhi Call Girls Punjabi Bagh 9711199171 ☎✔👌✔ Whatsapp Hard And Sexy Vip Call
An Ontology for formal representation of Drug-Drug Interaction Knowledge
1. 6th International Biocuration Conference
April 7-10, 2013, Cambridge, UK
An Ontology for formal representation of Drug-Drug
Interaction Knowledge
María Herrero-Zazo, Isabel Segura-Bedmar, Paloma Martínez
Computer Science Department, Universidad Carlos III de Madrid, Spain
{mhzazo, isegura, pmf}@inf.uc3m.es
Motivation:
A drug-drug interaction (DDI) occurs when one drug influences the level or activity of
another drug.
Drug Interaction databases are rarely complete => Medical literature is the most effective
source for the detection of drug interactions.
The process that leads to DDI may involve pharmacokinetics (ADME) and
pharmacodynamic processes. Therefore, the biological mechanisms underlying DDIs include
interactions with metabolic enzymes, protein transports and drug targets.
The formal representation of this knowledge can be useful for: data annotation tasks, drug
discovery, in silico prediction of DDIs and early detection in clinical practice, improvement of
Clinical Decision Support Systems, development of signal detection methods in
Pharmacovigilance, integration in Electronic Medical Records, etc.
The Drug-Drug Interactions Ontology
ACKNOWLEDGMENTS:
This work was supported by the Regional Government of Madrid under the Research Network MA2VICMR [S2009/TIC-1542] and by the
Spanish Ministry of Education under the project MULTIMEDICA [TIN2010-20644-C03-01]
We thank the team at the Humboldt-Universitaet zu Berlin for making available a visualization of the DDI corpus using Stav:
http://http://corpora.informatik.hu-berlin.de/
https://github.com/TsujiiLaboratory/stav
The DDI Corpus
• The largest corpus for DDI Extraction
•Two different types of documents: DrugBank drug interaction fields and MedLine
abstracts.
•Annotation of different types of pharmacological substances and PK and PD DDIs.
•Manually annotated by two annotators relying in specific annotation guidelines.
•Measurement of Inter-Annotator Agreement (IAA).
•Publicly available.
•The DDI Corpus is being used in the DDI Extraction 2013 task, for the recognition of drug
names and extraction of drug-drug interactions in biomedical literature.
Schedule Specification
Knowledge
Acquisition
Conceptualization Implementation Integration Evaluation
Knowledge
acquisition
The DDI
corpus
MANUAL
Annotation process
Identification of main
concepts, attributes
and relationships
COMPUTER BASED
EXTRACTION
ELICITATION
Modeling the domain
REFINEMENT
Integration between granularities
Connection with other ontologies
CODIFICATION
IRMORPHOLOGICAL
PRODUCTIVITY
The
DDI
corpus
Study, comparison
and evaluation
1ST
Prototype Life Cycle
SNOMED CT®
National Drug File (NDF)
WHO-ART
The PK Ontology
The DIO Ontology
The DDI Ontology
RxNorm Ontology
National Drug Data File (NDDF)
COSTART
Pathway ontologyThe Protein Ontology (PRO)
Some Related Ontologies
Total
Documents
DrugBank 792
MedLine 233
Sentences
DrugBank 6,795
MedLine 2,147
Entities 18,502
DDIs Relationships 5,028
DDI
Effect
Management
PD_mechanism
Mechanism
Drug
has_precipitant has_object
produces
is_avoided_by
is_produced_by
PK_mechanism
is_a is_a
A preliminary Semantic Network Representing DDI Knowledge Structures