Robustness and Evolvability in Living Systems. Andreas Wagner. Princeton University Press, Princeton, NJ, 2005. 408 pp., illus. $49.50 (ISBN 0691122407 cloth).
If you are an active member of the biological scientific community, you cannot have escaped noticing: Everything seems to be robust these days. The genetic code is a robust encoding of amino acids into codons, RNA molecules are robust to point mutations, proteins are robust to translation errors, developmental pathways are robust to environmental or genetic disturbances, metabolic networks are robust to changes in enzyme efficiency…the list goes on and on. While not all of these ideas are new (e.g., the idea of robustness in organismal development goes back to Conrad Hal Waddington's work in the 1950s), the amount of research devoted to the robustness of various biological systems has exploded in recent years. However, there is a danger in using a concept such as robustness, in that it can be so generic and widely applicable that its use may contribute little to the understanding of specific biological systems. For example, it is not clear a priori that a robust RNA molecule that can fold correctly despite nucleotide substitutions has anything to do with a robust metabolic network that continues to function after gene deletion.
Enter Andreas Wagner's book on robustness. Wagner's book has two aims, to review the current knowledge on robustness and to identify common principles that cause biological systems to be robust. The book is divided into four parts. The first two parts, “Robustness Below the Gene Level” and “Robustness Above the Gene Level,” provide a comprehensive review of all things robust. Wagner begins with nucleotides themselves; talks about the genetic code, RNA, and protein molecules; and moves to successively larger scales, including metabolic pathways, phenotypic traits, and body plans of whole organisms. Each chapter is well researched, provides a large number of useful references, and gives a concise and readable introduction to its subject area.
The third part,“Common Principles,” tries to bring some order into the bewildering array of robust systems introduced in the first two parts. One of its central themes is the concept of a neutral space, defined as “a collection of equivalent solutions to the same biological problem” (p. 195). The main question raised in this section, to my mind, is whether robustness arises as a by-product of other constraints (the rules of biochemistry dictate, for example, that in metabolic pathways the overall metabolic flux is not strongly dependent on the exact enzymatic activity of each participating enzyme) or whether there is a selective pressure to increase a system's robustness to mutations (as in van Nimwegen and colleagues' model of evolution of mutational robustness, introduced in Proceedings of the National Academy of Sciences in June 1999). Wagner addresses this question in chapter 16, where he discusses a variety of mathematical models of evolution that lead to robust systems. I found the third part of the book somewhat less satisfying than the first two parts, not because Wagner doesn't do a good job in describing the various models he presents (he does), but because ultimately it doesn't seem possible to construct a unified theory: “These differences suggest that there may be no fundamental theory of how robustness evolves, if such a theory is required to take into account the different architectures of biological systems” (p. 268).
The book closes with a short fourth part, “Robustness Beyond the Organism.” Consisting of only two chapters, it lists some more examples of robust systems beyond the main scope of the book and draws parallels to the material discussed in the previous chapters. The first of the two chapters discusses robustness that arises through self-organization. The main example in this chapter is the self-organization of ecosystem communities into assemblies that are resistant to invasion. The second chapter discusses robustness in man-made systems, such as telecommunication networks and evolved hardware.
Should you read this book? Wagner himself seems to suggest that the book is primarily for those new to the field of robustness, as he writes in the introduction: “If you are a specialist who already knows some or most of the literature in this field, much of this book will not be news to you” (p. 10). However, I would argue that this book is invaluable to everybody interested in robustness, for two reasons. First, even if you are a specialist, you will probably not be an expert in all the different areas Wagner touches on. Second, the book serves not only as an excellent introduction into a wide range of literature on robustness but also as a handy work of reference, with a bibliography totaling over 600 entries. Need some key references on the robustness of the genetic code? Just skim through chapter 3, and you have them. I predict that for many years to come, Wagner's book will be the bibliographic reference work of choice for research on robustness.