ABSTRACT
This chapter describes a fairly simple simulation that is conducted based on a mouse genetic data experiment, and provides an overview of some of the issues that arise. It considers available gene network software, from both Bayesian and frequentist perspectives, and explains the existing correlation structures in the data as a gene network. With the advent of genomics and much more detailed genetic information, there has been a renaissance of statistical applications in genetics. Patterns in gene expression that are related to the onset of cancer are an area of great research interest. The data analysis has assumptions regarding underlying linearity in the relationships between variables not in regard to a potential likelihood or prior density. With the initial data analytic approaches, most of the higher dimensional models, from a frequentist or Bayesian likelihood perspective, assume basic linearity in gene expression on the employed scale.
