Promoting integration across diverse studies of environmental health: Development of a children's health and exposure ontology

摘要

Collective-science initiatives rely on the investigators' ability to harmonize data and methodologies across diverse studies in order to promote integration. The Children's Health Exposure Analysis Resource (CHEAR) provides the infrastructure to promote the inclusion of exposure science within diverse studies of children's health. In order to support the CHEAR Data Center's mission of providing support for the analysis, interpretation, curation and reuse of data within the CHEAR network, the CHEAR ontology was developed to provide a common vocabulary with unambiguous definitions of concepts and variables across studies. By integrating with existing ontologies and supplementing terms unique to exposure science and epidemiology, the CHEAR ontology supports chemical, biological and epidemiologic measurements that comprise the majority of data within environmental health studies. This includes distinctive features of environmental health research such as repeated measures of environmental biomarkers across time and parent-child linkages within study populations. Currently, the CHEAR ontology interfaces with thirteen foundational ontologies and contains 232 chemical analytes, 165 epidemiologic attributes, 21 biologic sample types, 15 classes of laboratory assays, and 150 laboratory instruments. Customizable templates have been created for researchers to accurately describe their data in a way that facilitates their semantic integration with other environmental health studies and existing ontologies. The template approach allows a wide range of users to input content while supporting dynamic growth of the ontology. The creation of this ontology will facilitate both the pooling of study data and the construction of searchable data warehouses with epidemiologic and environmental exposure data. The CHEAR ontology can be reused and expanded as needed to specify, curate and search data from studies within the broader fields of exposure science and environmental epidemiology.