U.Alon, An Introduction to Systems Biology:Design Principles of Biological Circuits Chapman and Hall/CRC 2007
Biology is complicated but there are simple underlying principles: this is the message of Uri Alon’s timely book giving an introduction to Systems Biology. In this he is following in a distinguished tradition of the rationalization of Biology starting with the molecular revolution, the defining Watson-Crick moment, and continuing in the forceful popular expositions of Richard Dawkins.
Alon is a physicist by training and this reflects in his approach: using the art of “toy models” to capture something of essence about a system (more on this below). In a series of chapters he demonstrates the success of this approach.
The first part of the book is devoted to exploring regulatory network motifs. Consider the graph whose nodes are genes and edges represent regulatory interactions for instance the protein coded for by one gene activates or represses the gene for producing another protein. A motif is a subgraph in this network that is significantly overrepresented statistically, as compared to a random network. The hypothesis is that such a significantly over-represented subgraph cannot occur by chance but must have evolved by selective pressure.Hence it must serve some definite crucial function-. Alon examines a number of such motifs exploring their dynamics through simple ordinary differential equation models. Amazingly, one discovers this way, that a number of familiar engineering principles are implemented via these motifs. For example, control theorists use the concept of integral feedback to maintain stability of systems such as heat controllers, and – lo and behold! – the same principle is operating in biological systems!
In the second part of the book, Alon concentrates o one of the most important and remarkable properties of biological systems, namely their robustness i.e. ability to adapt successfully and reliably to many different situations. He illustrates a number of principles of robust design , again using quantitative toy models. A beautiful example he uses is, for instance to ask the question:which genes have positive regulation and which negative? And is there a good reason for the choice?
One misgiving I have is about how biologists would take to these toy models, especially once the going gets somewhat more difficult. To his credit, Alon never needs much beyond the simplest differential equations, but even this, I fear, may be outside the remit of a typical biologist today.
A solution I propose, as a computer scientist, is to use one of the many simulation and modelling tools available today, to shield the biological student from the daunting differential equations. At least for a start, one could build the toy model in such a system with the underlying mathematics hidden and allow the user to play around and experiment with the model, thus aining an understanding of its properties. At a later stage, the hood may be opened for those who dare, to reveal the underlying model. Another advantage is that this way, one can use different kinds of underlying models either deterministic and stochastic, and see the differences in behaviour, which are sometimes significant.
But the greatest advantage of this approach is that it brings to bear the more powerful and flexible computational approaches on investigating biological models. Toy models are great to understand principles and Alon is happy to point out that all the models in the book can be solved on the blackboard or on a small piece of paper but this is really the exception than the rule – even the simplest of toy models can sometimes be forbiddingly difficult to analyse precisely using ODEs. Computational tools can thus greatly expand the scope of use of toy models.
Biology may be complicated but I fear there is no hope for us to understand it unless we take the stand that Alon advocates and demonstrates so successfully in this book: that there are simple underlying principles of design.
