University at Buffalo
Assistant Professor, Department of Biostatistics
Graphical models have proven to be a valuable tool for connecting genotypes and phenotypes. Structural learning of phenotype-genotype networks has received considerable attention in the post-genome era. In recent years, a dozen different methods have emerged for network inference, which leverage natural variation that arises in certain genetic populations. The structure of the network itself can be used to form hypotheses based on the inferred direct and indirect network relationships, but represents a premature endpoint to the graphical analyses. In this work, we extend this endpoint. We examine the unexplored problem of perturbing a given network structure and quantifying the system-wide effects on the network in a node-wise manner. We leverage belief propagation methods in Conditional Gaussian Bayesian Networks (CG-BNs), in order to absorb and propagate phenotypic evidence through the network. We show that the modeling assumptions adopted for genotype-phenotype networks represent an important sub-class of CG-BNs, which possess properties that ensure exact inference in the propagation scheme. Applications to kidney and skin cancer expression Quantitative Trait Loci (eQTL) data will be presented. We demonstrate how these predicted system-wide effects can be examined in connection with estimated class probabilities for covariates of interest, e.g., cancer status. Despite the uncertainty in the network structure, we demonstrate the system-wide predictions are stable across an ensemble of highly likely networks. A software package, BayesNetBP, which implements our approach, has been developed in the R programming language and is available on the Comprehensive R Archive Network.
Rachael Hageman Blair, PhD joined the department of Biostatistics in the Fall of 2011. She graduated from Case Western Reserve University in 2007 with a PhD in Mathematics. Her graduate work focused on the development of integrated numerical and statistical methods for mathematical models of metabolic systems. Blair's postdoctoral training was in statistical genetics at the Jackson Laboratory. The focus of her research was in methodology development for genotype-phenotype inference with applications to complex traits in mouse genetics. In addition to her methodological research, she has a number of collaborative publications, which involve the application of bioinformatics and data mining techniques to analyze high-throughput data. Her primary research interests center around modeling complex social and biological systems.