Elements of Computational Systems Biology

2

MODELS IN SYSTEMS BIOLOGY: THE PARAMETER PROBLEM AND THE MEANINGS OF ROBUSTNESS

Jeremy Gunawardena

Department of Systems Biology, Harvard Medical School, Boston, MA, USA

 

With four parameters I can fit an elephant and with five I can make him wiggle his trunk.—told by Enrico Fermi to Freeman Dyson and attributed to John von Neumann [1].

2.1 INTRODUCTION

I coteach a graduate course at Harvard called An Introduction to Systems Biology. It covers some of the mathematical methods used to build mechanistic models of molecular and cellular systems. Beginning students tend to ask two kinds of questions. Those with a biological background say “Why do I need to use mathematical models? What can they tell me that conventional biological methods cannot?”, while those from the physical sciences (mathematics, physics, and engineering) or computer science say “I know how to model. Why is biology any different from physics or engineering?” Broadly speaking, everyone wants to know, from very different perspectives, “How do I do systems biology?” Students are usually under the misapprehension that the person standing in front of them knows the answers to such questions. In my case, I was only marginally less ignorant than the students themselves. It was their curiosity and skepticism, along with a realization that the field lacks a shared foundation for discussing such questions, that forced me to think more deeply about the issues.

This paper is the first of at least two in which I review some ...