Re: Recycling, Rosennean Style...

[Leo Caves <***>]

Judith said:
> But the idea of molecular reconfiguration, as a form of recycling, is 
> worth exploring, it seems to me. This deals with simple organization 
> and should be perfectly "do-able" with current scientific modes and 
> methods. Anybody have a friend who's a theoretical chemist?

Some efforts in this direction are termed "bioremediation" - co-opting, 
or "evolving" microorganisms (and thereby their complex metabolic 
networks) to render undesirable waste products (such as radionuclides, 
toxins) safer (or ideally "inert")

The problem with transforming (or breaking down) undesirable (not 
always large) molecules to safer entities is that it may require 
significant (free) energy to do so or the kinetics of degradation 
[half-lives] may be too long, to be of any practical benefit.
Solutions:
1/ Supply the energy to overcome the barriers / speed the process
Applying lots of heat in furnaces, assuming that one breaks down to 
near elemental products.  A problem being that at such high 
temperatures one also expands the width of the product distribution and 
you might get (nasty) surprises.  The chemical universe is *VAST*, and 
as such are immune systems cannot cope with the scale of potentialities 
of such non-natural products - hence issues of toxicity and 
bioaccumulation.

2/ Lower the barriers to transformation
Enzymes act as catalysts to chemical reactions by providing a local 
microenvironment that lowers the energetic cost of a chemical 
transformation thereby accelerating the reaction.   Reactions performed 
by single enzymes tend to be rather specific; however metabolic 
networks (the graph of possible enzymatic reactions) in an organism may 
be extensive and complex (having scale-free characteristics as you 
know) and thus provides myriad possible transformation routes according 
to its evolutionary adaptation.
A naturally occurring bug may do the required remediation 
transformation for you, alternatively you can use processes of 
"directed evolution" to optimise a desired outcome (or promote new ones 
- though harder!).

A quick google reveals a US DOE funded bioremediation research 
programme at Lawrence Berkley National Labs:
http://www.lbl.gov/NABIR/generalinfo/intro.html

I agree that your more holistic view of "bioremediation", which would 
require (as you state) a more systemic approach.

A small comment: in software engineering it is considered good practice 
to provide a "destructor" function for every "constructor" of data or 
processes.   If you screw this up, you risk ill-defined, poor or even 
fatal consequences in your execution!  This process can be automated by 
providing your environment with a garbage collector - which recognises 
the importance for this complementary pair of functions, and takes it 
out of the hands of the feeble programmer and embeds it in the 
underlying run-time system.

Leo
(a computational chemist, now working in biology)

--
Leo Caves, Computational Biology, University of York
http://www.york.ac.uk/biology/staff/lsdc.htm