Biology Reference
In-Depth Information
even before it is tried if, as in
Figure 28.4
, the analytical phase of the exercise produced the
same sequence of module activation from examination of the development of two different
body structures. Just evoking modules is not enough to predict development; fine details,
including initial 'boundary' conditions and subtle differences in the working of a 'basic'
module, such as aggregation in one system and in another, matter too much. It could be
argued that the problem is that the exercise failed because it was trying to describe develop-
ment in terms of too few modules; and if we had 50 different versions of an aggregation
model, and similarly large numbers of all other modules, then it might be possible to succeed.
Indeed it might, but by invoking the existence of tens of subtle different versions of every-
thing even within the same organism, we have admitted that the idea of just a few simple
basic modules/integrons does not work for predictive modelling of real embryos. This
complete exercise, and the argument, can be run at any module scale frommolecules to social
groups.
)
The fact that organizing our understanding of morphogenesis around modules/integrons
does not yield a predictive model does not mean that the approach is useless. It is an excellent
aid to understanding, to discussion and to organizing information. It is, though, an over-
simplification and it should be used only by people willing to remember its limitations. As
George Box said, '
All models are wrong, but some models are useful
'
.
ARE MULT
ILAYERED MORPHOGENETIC SYSTEMS HIERA
RCHICAL?
One concept that is frequently mentioned in standard textbooks about the molecular cell
biology of development is that of 'hierarchical control'. 'Hierarchy' is a term that originates in
organized religion
A
and refers to a chain of command in which agents are assigned to levels,
and each level is in command of the levels below and is commanded by the levels above. In
many religions, junior priests are subordinate to high priests, and other social structures such
as the army have adopted a similar hierarchical structure in which people of low rank are
subordinate to those of higher rank. Like modularity, the concept of hierarchy is both useful
and dangerous in the context of biology, and in the context of morphogenesis its dangers
probably outweigh usefulness and it should be dropped.
The idea of hierarchical control entered molecular biology mainly through research into
the molecular genetics of simple prokaryotic systems such as bacteriophage
and the
lac
operon.
25
In these systems, the activity of one gene (for example, the cI repressor of bacterio-
phage
l
) controls the expression of a number of other genes and is therefore regarded as
a 'master regulator', at a superior position in a hierarchy of control. The hierarchical model
is useful for simple and relatively inflexible systems such as viruses, but is less so for more
complicated and flexible mechanisms. The problem is that the feedback in these systems, the
same feedback that gives them the property of adaptive self-organization, breaks
l
)
A similar exercise can be done with literature: Georges Polti famously asserted
24
that all stories in litera-
ture and drama are combinations of 36 basic dramatic situations (
modules). It is easy to analyse works of
literature into these plot lines, but not so easy to shuffle plot modules and come up with a new great work of
literature. Again, many different stories follow the same sequence of modules.
A
From the Greek
hierarkh
ˆ
s
z
¼
high priest.
