Variable Selection in Semiparametric Models by Subsampling

• Activity Number: 593
• Type: Invited
• Date/Time: Thursday, August 4, 2011 : 8:30 AM to 10:20 AM
• Sponsor: IMS
• Author(s): Wenxuan Zhong*+
• Companies: University of Illinois at Urbana-Champaign
• Keywords: variable selection ; dimension reduction ; semiparametric model ; RNA-seq

Abstract

With the rapid development of next generation sequencing technologies, RNA-Seq has become a popular method for transcriptome analysis. After mapping to the genome and/or to the reference transcripts, RNA-Seq data can be summarized by a sequence of read counts. Calculating gene expression from RNA-Seq data is based on these counts. This new frontier also raises significant statistical challenges for researchers who are unable to access to large computing clusters, because there is a lack of effective and efficient statistical tools for handling the super-large dataset. It calls upon an urgent need for effective statistical modeling and computing methods, because many classical models and methods derived for moderate large data are no longer applicable. To facilitate statistical modeling using current computing resources, we present, in this talk, a subsampling method for variable selection in semiparametric models of RNA-seq.