INTELLIGENT DEVELOPMENT 145genic proteins). Most of them are derived from knockout studies and de-
scribe consequences of chemically inhibiting a signal. Th ey have pressed
home the need for a new perspective on the causes of early diff erentia-
tion and development.Dynamical Rather Than Deterministic Pro cesses
Wolpert’s gene- centered model was already being challenged by Brian
Goodwin in the early 1980s. Goodwin spoke of a more “self- organized
entity” born out of the “relational order” among active entities. He sug-
gested that the genes play only a secondary role in establishing the pat-
terns of development. Th at has been confi rmed in recent molecular
biological studies. For example, enumerating all the genes turned on in
a par tic u lar location does not predict the developmental outcomes. Th e
latter are determined by the global physiological state of the cell in the
morphogen gradients, all dynamically self- organ izing in attractor land-
scapes, as described in chapter 4.
Take, for example, Wnt signaling proteins. Th ese are a large family
of nineteen proteins helping coordinate a daunting complexity of signal-
ing regulation and function in development: cell fate, cell motility, body
polarity and axis formation, stem cell renewal, organ formation, and
others. But Wnt proteins are themselves tightly regulated in feedfor-
ward and feedback loops. As Yuko Komiya and Raymond Habas noted
in a review, Wnt proteins and their antagonists “are exquisitely restricted
both temporally and spatially during development.”^5 Th ey are heavi ly
modifi ed prior to transport and release into the extracellular milieu.
Th en their activity levels, their shape, and degree of binding to target cell
membranes are regulated by a number of co- factors, including other
morphogens.
Wnt signaling is just one of a host of morphogenic pathways. Th e pro-
fusion of them explains why many embryologists and developmental bi-
ologists are now resorting to mathematical modeling of developmental
systems. As described in chapter 4, the form and drift of interactions
among large collections of signaling networks, TFs, RNAs, and so on, is
best described in terms of attractor states.^6 It may be remembered from
chapter 4 that one of the characteristics of such systems is that they are
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:53:16 UTC