Information Technology Reference
In-Depth Information
by Thakar and Albert introduces the reader to Boolean modeling methods, showing
how these can be used to infer causal relationships from expression data, and also
to analyze the dynamics of systems whose network of interactions is known.
The global dynamics of gene regulatory networks must be robust in order to
guarantee their stability under a broad range of external conditions. The eect of
noise on the dynamics of Boolean networks is investigated in the Chapter by Daz-
Guilera and Alvarez-Buylla. In this case, the noise simulates errors and external
stimuli aecting the transmission of signals in real living processes. The authors
focus on an experimentally grounded gene regulatory network that describes the
interactions required for primordial cell fate determination during early stages of
ower development. This is a very instructive example, since it shows how the noise
maximizes the capacity of the system to explore the state space, while at the same
time the network is able to retain the observed steady states under dierent noisy
regimes.
The Chapter by Bongini and Casetti is about protein folding, one of the most
fundamental and challenging open questions in molecular biology. The core of
the protein folding is to understand how the information contained in a sequence
of aminoacids is translated into the three-dimensional native structure of a pro-
tein. And to clarify why all the natural selected sequences of aminoacids fold to
a uniquely determined native state, while a generic polypeptide does not. With
these questions in mind, the authors describe two dierent strategies to analyze
the high-dimensional energy landscape of model proteins. The rst approach is
essentially topological in character, and amounts to dene a network whose nodes
are the minima of the potential energy and whose edges are the saddles connecting
them. The second approach is based on the denition of global geometric quantities
that characterize the folding landscape as a whole. The reader will lear that both
methods can give interesting information on the dierences between the landscapes
of protein-like systems and those of generic polymers.
A simplied representation of the potential energy function of a protein near
equilibrium can be obtained, at low computational cost, by using the so-called
elastic network models (ENMs). In ENMs, the aminoacids are the nodes of the
network, represented as point particles in three dimensions (3D), while the edges
of the network are springs joining the nodes, representing harmonic restraints on
displacements from the equilibrium structure. The attractive feature of ENMs
is that they provide an intuitive and quantitative description of the behavior near
equilibrium. Furthermore, the few parameters used in ENMs can be easily adjusted,
giving uncommon adaptability to the method. There are, however, some limitations.
Although ENMs robustly predict collective global motions, they do not provide
reliable descriptions of local motions. Also, the harmonic approximation requires
a potential minimum, limiting the utility of ENMs for modeling non-equilibrium
dynamics. In the Chapter by Lezon et al. the theory behind the ENM and some
of its extensions are reviewed. Finally, some recent applications, mainly focusing
