Biology Reference
In-Depth Information
Chapter 13
Genotype Networks and Evolutionary
Innovations in Biological Systems
Andreas Wagner
University of Zurich, Institute of Evolutionary Biology and Environmental Studies, Y27-J-54, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland
Chapter Outline
Introduction
251
The Diversity of Neighborhoods in Genotype Space
258
Metabolic Networks and Their Innovations
252
Genotype Networks and Their Diverse Neighborhoods Can
Help Explain the Origin of New Phenotypes
Regulatory Circuits and Their Innovations
252
259
Macromolecules and Their Innovations
253
Robustness, Genotype Networks and Environmental
Change
Towards A Systematic Understanding of Innovation
254
260
Genotype Spaces and the Phenotypes Therein
254
Conclusions and Future Challenges
261
Genotype Networks
256
References
261
INTRODUCTION
How new traits originate in life is a question that has occu-
pied evolutionary biologists since Darwin's time. This holds
especially for traits that are evolutionary innovations, i.e.,
qualitatively new features that benefit their carrier
[1,2]
.
About this origin, the geneticist de Vries said in 1904 that
Darwin's theory can explain the survival of the fittest but not
its arrival
[3]
. Today, more than a century later, the bio-
logical literature contains many well-studied examples
of innovations, fascinating case studies of natural history
[1,4
structure and function of its macromolecules, such as
protein and RNA molecules.
New phenotypes often arise through mutations that alter
an organism's genotype. Therefore, understanding pheno-
typic variability requires understanding how genotypic
change translates into phenotypic change. Ideally, experi-
mentation should provide this understanding
[11,12]
.
However, a systematic understanding of the relationship
between genotype and phenotype requires the analysis of
thousands if not millions of different genotypes and their
phenotypes. It is beyond reach of current experimental
technologies for most systems. An alternative is to use
existing comparative data about genotypes and their
phenotypes. The necessary information is available only for
a few kinds of system, for example proteins, where the
structure and function of tens of thousands of proteins are
available. In most other systems computational modeling of
phenotypes will be essential for the foreseeable future.
Fortunately, the tools of systems biology have allowed us to
make great strides in such modeling. For example, within
the last 15 years it has become possible to computationally
predict the biosynthetic phenotypes of enormously
complex metabolic networks comprising hundreds of
enzymatic reactions
[13,14]
. The analyses reviewed in this
chapter use such computational approaches, as well as
comparative data and experimentation. Taken together,
10]
. However, we still know little about any principles
that might underlie the origin of innovations, other than the
well-worn notion that a combination of mutation and natural
selection may be necessary. We do not even know whether
such principles exist. De Vries' statement makes clear that
such principles would be principles of how biological
systems bring forth novel and beneficial phenotypes. They
would be principles of phenotypic variability.
To understand the origins of new phenotypes one needs
to understand the relationship between genotype and
phenotype. The genotype is the totality of an organism's
genetic material. The phenotype is any other observable
characteristic. It includes the morphology and behavior of
complex organisms, the structure of cells, the expression
pattern of genes and proteins, the biosynthetic abilities of
an organism's metabolism, and the three-dimensional
e
