4 Responses to “Design principles of a bacterial signalling network”

1. Krauze Says:
   November 23rd, 2005 at 4:54 pm

   Hi Bipod,

   That reminds me of this quote, from a book I read a while ago:

   "Perhaps the most important feature of the development of complex organisms is one that permits evolutionary change. It is a property of development that has not been sharply identified and defined, and here I wish to do so. I will call the phenomenon a "gene net" and by this I mean a gathering, a grouping of a network of gene actions and their products into discrete units during the course of development. … The gene nets must become isolated from one another and proceed independently. It is not total independence; they still must keep in touch with one another, for this is the way pattern formation is achieved. But the signals between them are likely to be few and quite specific. … (An interesting aside: It has been pointed out to me by my former colleague Jon Seger that a properly structured "modular" computer program is put together in a way that is analogous to the gene nets postulated here.)"

   John Tyler Bonner, The Evolution of Complexity by Means of Natural Selection (Princeton University Press, 1988), pp. 174-5. My emphasis.

Comment by Krauze — November 23, 2005 @ 4:54 pm

2. AdR Says:
   November 23rd, 2005 at 6:36 pm

   The gene nets must become isolated from one another and proceed independently. It is not total independence; they still must keep in touch with one another, for this is the way pattern formation…

   …properly structured 'modular' computer program is put together in a way that is analogous to the gene nets postulated here

   That sounds much like 'design-by-contract'.

   So studying the underlying design, one can deduct the gene organization. I call that a succesful design approach to evolution.

Comment by AdR — November 23, 2005 @ 6:36 pm

3. DonaldM Says:
   November 23rd, 2005 at 9:18 pm

   I call that a succesful design approach to evolution.

   Could you clarify what you mean here?

   Thanks!

Comment by DonaldM — November 23, 2005 @ 9:18 pm

4. AdR Says:
   November 24th, 2005 at 5:14 am

   Apparently, they found out that the gene nets show a striking similarity to the design patterns that software architects would use. The best (and scientific) approach would then be that you start modelling the gene nets on software methodologies. In that case, you have a framework on which to model evolution. That's more than neodarwinism ever dreamed of.

   The question in this stage is whether it is a coincidence that the gene nets show similarities with design patterns. I'd say it is not, because the design patterns just define 'best practices' for the architecture of complex systems.

Comment by AdR — November 24, 2005 @ 5:14 am