Statistic Models in Biological Network Analysis

In the post-genomic era, genome-wide experiments have become commonplace and methods to interrogate such datasets are becomingly increasingly important. One approach to analyzing these large datasets is to model the experimental observations using a network approach. In a process known as (re)construction, inference, identification or reverse engineering, model parameters are fit to the data yielding a defined network that can then be analyzed to gain higher level insights into the complex molecular biology. Network biology is an expansive field and there are many different methods that are employed. In this survey paper we briefly review the frameworks of several popular modeling methods including logical models, boolean networks, Bayesian networks, weighted correlation networks, differential equations, and some basic network analysis concepts.