You are here:
Integration into Big Data: First Steps to Support Reuse of Comprehensive Toxicity Model Modules (SOT)
Watford, S., J. Edwards, I. Rusyn, D. Reif, R. Judson, AND M. Martin. Integration into Big Data: First Steps to Support Reuse of Comprehensive Toxicity Model Modules (SOT). Presented at SOT Annual Meeting, San Diego, CA, March 22 - 26, 2015. https://doi.org/10.23645/epacomptox.5178712
Present poster at SOT
Data surrounding the needs of human disease and toxicity modeling are largely siloed limiting the ability to extend and reuse modules across knowledge domains. Using an infrastructure that supports integration across knowledge domains (animal toxicology, high-throughput screening, genomics, proteomics, disease, exposure, product use, chemistry, etc.) increases the ability to evaluate, extend and expand models. For example, type II diabetes is a metabolic disorder caused and influenced by a combination of genetics, lifestyle and environment. In order to quantify the contribution of each factor and related confounders (e.g., diagnosis, screening, and treatment), the modeling framework relies on the ability to systematically access information across many knowledge domains to more accurately resolve the uncertainty resulting from the complexity within and across each factor. A first step to developing an integrated system was to develop an object model (i.e., a conceptual representation of each knowledge domain; ontologies) to resolve data redundancy and granularity issues from the complexity of the data. The advantage of an object model over siloed databases was the ability to confidently link and merge previously disconnected datasets. The current object model enables the modular development of systems capable of providing an extensible framework for building a more comprehensive human disease model. This abstract does not necessarily reflect US EPA policy.