An Application of Penalized Regression to Biomarker Selection in a Cohort of Bangladesh Children

• Activity Number: 193
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Miao Lu* and Jianhui Zhou and Caitlin Naylor and William A. Petri and Jennie Z. Ma
• Companies: University of Virginia and University of Virginia and University of Virginia and University of Virginia and University of Virginia
• Keywords: Malnutrition ; Vaccine Failure ; Biomarker Selection ; Group SCAD

Abstract

Environmental Enteropathy is a subclinical condition caused by constant fecal-oral contamination and resulting in blunting of intestinal villi and intestinal inflammation. We seek to evaluate the association between non-invasive biomarkers and vaccine failure and predict development of malnutrition in a cohort of Bangladesh children.

Regression analysis were performed, and the results were summarized in a heat map plot with False Discovery Rate adjustment. Cluster analysis of variables revealed the relationship among biomarkers. Furthermore, we applied different variable selection methods with different criteria to select a subset of informative biomarkers. The selected biomarkers for malnutrition and vaccine failure in our study are gut damage, inflammation, maternal factors.