Chapter 5
Cells in Motion
5.1 Why Such a Tangle?
The first open question posed by Freeman Dyson (2004) is: Why is life so complicated?” This is, indeed, a great inconvenience for a physicist, a habitual model
builder. Look at the two networks shown in Fig. 5.1. On the left is a model of the
metabolic network of the lowly bacterium E. coli, and on the right, one of the sociograms pioneered by Jacob Moreno (1934). There is a difference in the graphics:
on the left, all arrows are directed from one protein to another one that it catalyzes,
while on the right, many arrows indicating human interactions or influences are
double-headed. Nevertheless, the graph on the left also contains closed autocatalytic
loops. If we don’t care about the meaning of the letters in the circles, we are left with
an abstract graph of the kind studied by the theory of networks, which aspires to universality and does not care about meanings either. Yet, the chemical nature of any
metabolic network is unique. Proteins, its nodes, have specific functions depending
on their ability to recognize and bind other molecules, ligands. This recognition depends on specific binding sites and the specific three-dimensional configuration of
the protein, and the variety of interactions among proteins far exceeds whatever we
might encounter in other networks.
A network of Facebook “friends”, or airline connections, or metabolism in the
human body cannot be drawn on a sheet of paper. The first two can be, and certainly are, fed into a computer with a sufficiently monstrous memory, but the last
one, though perhaps containing fewer elements and links, cannot be, because it is
unknown. One can compile useful statistics for large networks, either abstract or
specific, like the two I have just mentioned, but the statistics of a metabolic network
is good only for mathematicians to play with. Here we need specifics, chemical
structures, configurations of proteins, even reaction rates, which are rarely available
even in the simplest cases. How many millions of working hours, how many teradollars have been invested in studies of metabolic circuits involving the production
of a single protein in a single organism, often just a “model animal”, like the fly
Drosophila or the worm C elegans, or even the model microbe E. coli? There is
63
© Springer Nature Switzerland AG 2020
L. Pismen, Morphogenesis Deconstructed, The Frontiers Collection,
https://doi.org/10.1007/978-3-030-36814-2_5
Cells in Motion
5.1 Why Such a Tangle?
The first open question posed by Freeman Dyson (2004) is: Why is life so complicated?” This is, indeed, a great inconvenience for a physicist, a habitual model
builder. Look at the two networks shown in Fig. 5.1. On the left is a model of the
metabolic network of the lowly bacterium E. coli, and on the right, one of the sociograms pioneered by Jacob Moreno (1934). There is a difference in the graphics:
on the left, all arrows are directed from one protein to another one that it catalyzes,
while on the right, many arrows indicating human interactions or influences are
double-headed. Nevertheless, the graph on the left also contains closed autocatalytic
loops. If we don’t care about the meaning of the letters in the circles, we are left with
an abstract graph of the kind studied by the theory of networks, which aspires to universality and does not care about meanings either. Yet, the chemical nature of any
metabolic network is unique. Proteins, its nodes, have specific functions depending
on their ability to recognize and bind other molecules, ligands. This recognition depends on specific binding sites and the specific three-dimensional configuration of
the protein, and the variety of interactions among proteins far exceeds whatever we
might encounter in other networks.
A network of Facebook “friends”, or airline connections, or metabolism in the
human body cannot be drawn on a sheet of paper. The first two can be, and certainly are, fed into a computer with a sufficiently monstrous memory, but the last
one, though perhaps containing fewer elements and links, cannot be, because it is
unknown. One can compile useful statistics for large networks, either abstract or
specific, like the two I have just mentioned, but the statistics of a metabolic network
is good only for mathematicians to play with. Here we need specifics, chemical
structures, configurations of proteins, even reaction rates, which are rarely available
even in the simplest cases. How many millions of working hours, how many teradollars have been invested in studies of metabolic circuits involving the production
of a single protein in a single organism, often just a “model animal”, like the fly
Drosophila or the worm C elegans, or even the model microbe E. coli? There is
63
© Springer Nature Switzerland AG 2020
L. Pismen, Morphogenesis Deconstructed, The Frontiers Collection,
https://doi.org/10.1007/978-3-030-36814-2_5