Generating Consistent Spatio-Temporal Events of Exposure forTranslational Exposomic Research

摘要

The concept of the exposome include endogenous processes within the body, biological responses of adaptation to environment, and socio-behavioral factors beyond assessment of exposures. Exposomes therefore need to cross-link locations and occurrences of environmental influences, and resulting direct biological pathway alterations as well as mutagenic and epigenetic changes on the phenome. Data representing these events need to be associated with their limitations and uncertainties associated with using the data as exact quantifications of exposure. In order to address these needs, we are developing a metadata-driven exposomic data integration platform, OpenFurther (OF), as a part of the Utah Pediatric Research using Integrated Sensor Monitoring Systems (PRISMS) Informatics Ecosystem (Grant NIH NIBIB U54EB021973). Based on the methods used for data collection (measurement or observation), and the differences and uncertainties in environmental measurements and true exposures, we grouped exposomic events into six broad categories: Sensor, Clinical, Biospecimen-derived, Participant reported, Aggregates, and Computational Models. Each of these events have spatio-temporal coordinates that can be unbounded, dense, discrete, instants & intervals, indeterminate, and at finest granularity available with the source. OF generates these event from heterogeneous study data and stores them as documents in Big Data stores. These high-resolution exposome records are used for spatio-temporal abstraction, reasoning and Big Data analytics methods. They can also be transformed with minimal transformational information loss into analytical models to support a diverse set of translational research archetypes including sensor development, environmental chemistry, exposure pathways, mechanistic understanding of environmental species on the genome for precision medicine, pharmacodynamics studies, clinical trials, observational, comparative effectiveness, and epidemiological studies.