Title: "An Integrated Analysis of Genetic Variants and Co-Expressed Genes"
Speaker: Ms. Zhaoxia Yu, Division of Biostatistics, Department of Health Sciences Research, Mayo Clinic
Place: Electrical Engineering (EE) 270; March 27, 2007, Tuesday, 4:30pm

Abstract

An integrated biological network is of great importance for us to understand the nature of biological processes and to unravel the etiology of complex traits. The network can be reconstructed from different data such as genetic variants and gene expression data. One important question in this field is whether individual genetic variants (such as Single Nucleotide Polymorphisms, SNPs) or a set of them can explain at least part of the correlation between gene expression levels.

In this talk, I will describe the statistical methods I developed to test the difference between the marginal correlation and the partial correlation, conditional on a set of SNPs. I derived and validated two distinct but asymptotically equivalent tests. Based on these tests, I further proposed a two-stage approach to first select a set of covariates to condition on, and then to test whether the set decreases the partial correlation, relative to the marginal correlation. Those methods provide the basis to elucidate the genetic determinants of co-expressed genes. In addition, they are of general relevance and can be applied to other multivariate statistical problems.

I will also briefly discuss practical issues such as inferring missing genotypes, and model selection on a genome-wide scale.

This is a Statistics Bioinformatics COALESCE candidate interview (joint between Statistics and a department in the College of Agriculture)

Click here for a full schedule of BIOINFORMATICS SEMINARS, past and present.