Abstract:
|
In high throughput biological experiments, differential expression testing and set enrichment analysis are commonly used to summarize results and generate biologically meaningful hypotheses for further analysis and validation experiments. Conventional approaches to such testing often fail to account for individual variation in relevant background features, often due to a lack of pertinent data. These features are especially important in the context of metabolomics, where blood metabolite levels can react sensitively to changes in nutrient intake. Here we introduce a method for detecting differentially expressed metabolites and metabolic pathways which incorporates both a network based approach to detection, and adjusts for individual variation through the integration of nutrient intake data. We test our method on data from a controlled feeding study featuring two distinct diets. By integrating data on relevant background features and taking a network approach, our method yields conclusions which are more biologically relevant and have greater statistical significance than conventional approaches. Thus, the findings from our method may provide greater clarity for future research.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.