Five years ago we (George von Dassow, Ed Munro, Eli Meir, and I)  made
realistic mathematical/computer models of  two ancient and famous genetic networks
that act early in diverse embryos to establish spatial gene expression patterns
prefiguring the body plan.  Our models  revealed these networks to be astonishingly
robust. That is, they continue to make the correct pattern in the face of
thousand-fold variations in the strengths and functional forms of interactions
among participating genes.  After getting over my surprise that it was even
possible to design networks with such properties,  I now believe only networks
having this kind of robustness can be functionally heritable in polymorphic
populations.  What design features might endow genetic networks with the kind of
extreme robustness we found?  To probe for answers, I  wrote a computer program
that haphazardly generates  randomly connected networks made from about the same
number of biochemically sensible parts that constitute the segment polarity and
neurogenic networks.  We (Bjorn Millard, Ed Munro, and I) devised computer
algorithms that discover and catalog the stable expression patterns any network
can make, and, from all these, distills those patterns the network can make
robustly with respect to variations of its parameters.  The bottom line is that 19
out of 20 random networks that our program created (i.e. networks devoid of any
purposeful design whatever) could  make at least one, and usually many, complex
stable spatial expression patterns with the same high robustness that the real,
evolved, segment-polarity and neurogenic networks exhibit.  Several of these
random, non-designed networks turn out to be much more robust than either real
network. Only 1 out of 20 random networks is a loser; it did not make robustly any
interesting pattern at all.

      Our algorithms for finding patterns any network can stabilize show that it?s
possible to replace the differential equation models we used for each network (to
keep track of how the concentrations of all gene products change continuously with
time in a hexagonally packed sheet of many cells) by discrete logic models with
quantized far-apart concentrations, jumping abruptly from one concentration to
another at discrete time clicks.  Unfortunately, for any given network, there are
many different ways to do this -- different ways for different parameter values, no
way appropriate for all parameter values.

      Our in silico result that thoughtless, haphazard, non-design produces
networks whose robustness seems inspired begs questioning what else unintelligent
non-design might be capable of.