Accelerating Behavioral and Social Sciences through Ontology Development and Application: NIH Research Network Project (APRICOT).

Title: Advancing Prevention Research In Cancer through Ontology Tools (The APRICOT project) Abstract Progress in behavioral research in cancer prevention is hampered by imprecision and heterogeneity in research and how it is reported. Ontologies provide a way of addressing this barrier by promoting clarity,

coherence and interoperability across studies and data sets. We have developed the Behavior Change Intervention Ontology (BCIO) using ‘FAIR Guiding Principles’ that has been recognized as the most semantically rich ontology in its domain, and its user-informed method of development as particularly innovative. However, it does not cover all the key classes and relations needed. In

addition, our tools for using the ontology are at an early stage of development. We, the APRICOT team, propose three key areas of activity: (i) Extend the BCIO to cover additional details of two key behaviors (tobacco use and physical activity), methodological features of studies evaluating behavior

change interventions (including links between measures and constructs), and causal links between intervention components and mechanisms of action (processes of change). (ii) Integrate the BCIO and ontologies covering ‘real-world’ data such as patient records and census data to better address Social and Environmental

Determinants of Health (e.g., poverty, educational level, gender identity and sexual orientation, racial and ethnic group, and built environment). (iii) Develop tools and resources to build engagement with, and adoption of, the BCIO for, e.g., writing study reports and protocols, and synthesizing evidence for theory and

intervention development. To achieve this, we will foster and work with a cross-domain Community of Practice (CoP) with members from diverse backgrounds, perspectives and roles. Throughout the project we will follow FAIR Guiding Principles and established methodologies based on science working with our CoP as stakeholders to ensure that the ontology and its tools and resources are

developed and implemented in a way that promotes widespread adoption and meets the varied and evolving needs of users. We will work to ensure that the BCIO, its mapping to other ontologies and its tools and resources can be sustained and enhanced beyond the life of the project. We will use established methodologies to extend the BCIO and integrate (i.e., re-use classes) with, and where

this is not feasible to map (i.e., cross-reference) to, other ontologies. These include ‘top-down’ (expert- and user-led) and ‘bottom-up’ (searching reports and data) approaches to identifying priority classes and relations to add to the ontology and iterating the process of editing labels, definitions and other meta-data (including

cross-references) in close collaboration with our CoP to ensure that they meet its diverse needs. To upgrade and extend our tools and resources we will continue to adopt our established methodologies of ethical and efficient design used in generating the current version, again working closely with our CoP.