ABSTRACT
Many models of interest in the field of systems biology contain many unknown parameters, resulting in high-dimensional parameter spaces that must be characterized with a complex structure that is not well understood. Understanding such structure, and how it impacts the predictivity of models and the potential for the construction of alternative models, is an active area of research. This chapter begins by providing an overview of some of the mechanics of parameter estimation. Just as importantly, however, we also endeavor to consider parameter estimation within the broader context of modeling to describe how it relates to model construction, inference, selection, reduction, and analysis. The chapter closes with some thoughts about the somewhat fractured and multifaceted field of systems biology, highlighting how issues of parameterization and parameter estimation lie at the crossroads of different schools of thought. Even though the chapter mostly addresses mechanistic models of cellular processes, many of the concepts and techniques introduced here are broadly applicable to a wide range of models relevant not just to immunology and systems biology, but to other fields as well. In the field of immunology, this might include statistical models, or descriptions at other levels of biological resolution, such as models of the population dynamics of pathogens replicating within hosts, or spreading among hosts. Where possible, the author points out generalities and abstractions that are useful across different classes of models, while also noting some of the particular aspects that arise in analyzing complex cellular networks.
