The international, observational Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study, conducted 2000-2006, demonstrated independent, additive associations between maternal glucose and BMI during pregnancy with offspring birth weight and adiposity. Ongoing NIH-funded HAPO Metabolomics projects seek to characterize the maternal and fetal metabolic milieus in four ancestry groups underlying associations between these maternal and newborn phenotypes.
In this presentation, we will discuss statistical challenges in the analysis of metabolomics data. Using HAPO Metabolomics for illustrative examples, we will highlight analytic issues and proposed solutions including non-targeted metabolomics data normalization, metabolic network construction and subnetwork optimization, and differential network analysis. We will also describe strategies for constructing flexible analytic pipelines, including data visualization, that yield interpretable results for multidisciplinary teams
Audience
- Faculty/Staff
- Student
- Post Docs/Docs
- Graduate Students
Interest
- Academic (general)