robert karl stonjek

ON RECONSTRUCTION OF GENETIC CIRCUITS

The following points are made by D. Sprinzak and M.B. Elowitz (Nature 2005
438:443):

1) By taking apart an old clock, you could probably come up with a pretty
good guess at how it works. But a more concrete understanding of the clock
mechanism might be obtained by designing and building one's own clock out of
similar parts. Contemporary biology presents us with similar
reverse-engineering problems. For example, Drosophila cells contain a
circadian clock that oscillates with a 24-h rhythm and self-synchronizes to
the day/night cycle. Using genetic and biochemical techniques, researchers
have isolated genes and proteins involved in interlocked feedback loops of
gene expression[1,2] that are necessary for clock function. However, many
fundamental questions remain difficult to answer: what sets the period of
the oscillation, how does the clock operate reliably in diverse cellular
conditions, and what features of its design are responsible for its reliable
operation? To gain insight into such questions one could design and build
new clock circuits, using similar genes and proteins, and study their
dynamics in the organism. In fact, several synthetic genetic clocks have now
been constructed in bacteria[3-5]. These circuits are much simpler than the
Drosophila clock. They fail to operate as reliably, but they provide a proof
of principle for a synthetic approach to understanding genetic circuits.

2) As with the clock, many compelling biological questions center on how
interactions among genes and proteins, forming genetic circuits, give rise
to specific cellular functions. Genetic and biochemical techniques have
successfully identified many circuit components and their interactions.
However, in many cases knowledge about these components and their
interactions is not sufficient to explain the circuit mechanism. What is
missing from the circuit diagram? There are several possible deficiencies:
one is that the diagram may be incomplete -- interactions may have been
missed. The opposite problem also exists: the diagram may be too complete,
in the sense that it contains interactions that are not actively involved in
the process being investigated (for example, if one of the proteins is not
expressed under the relevant conditions). Another problem is ignorance of
the effective rules by which proteins and genes interact. For example, in
vivo values of kinetic parameters such as affinities, binding and
degradation rates, and so on, are generally unknown. Finally, the
intracellular environment is intrinsically "noisy", and small copy numbers
of molecular species limit the predictability of biochemical reactions.
Taken together, these problems reduce our confidence in the combined
understanding we get from perturbations, measurements, and mathematical
modelling.

Full Text at ScienceWeek
http://scienceweek.com/2005/sw051223-2.htm

Posted by
Robert Karl Stonjek