Latent Variable Methods for the Analysis of Genomic Data

• Activity Number: 142
• Type: Invited
• Date/Time: Monday, August 1, 2016 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): John D. Storey*
• Companies: Princeton
• Keywords: cross dimensional inference ; factor analysis ; surrogate variable analysis ; population structure ; confounding ; large-scale inference

Abstract

I will present flexible, robust, and computationally efficient latent variable methods that are effective in accounting for unobserved, systematic variation in genomic data. I will specifically focus on examples in gene expression studies, population genetics, and genome-wide association analyses.