Current genomics research indicates that statistical analysis based on individual genes may incur loss of information on the biological process under study. Better results can be derived from the analysis based on groups of genes, or gene networks. An informative characterization of a gene network is by the global Markov property, which can be inferred by the Gaussian graphical models (GGMs). In this talk, I will present two recent network inference problems with genomics data. The first one is in the case where, in addition to gene expression data, other types of data are collected from the same individuals. The problem is how to make use of this additional information when constructing graphical models for gene networks. The second one is in the case where gene expression data itself are collected under multiple tissues/conditions and the problem is how to jointly estimate multiple GGMs to increase power. After understanding these two problems, I will present new approaches for handling these problems as well as real examples.
<a href="http://www.stat.purdue.edu/~chunh/ ">PURDUE STAT</a>
Type of Event: