About the GM-microbes or plants. No, I’m not categorically against them.
However, the current venture capitalist driven proprietary model
with strong pressure to get profitable soon, and all the assorted
information hoarding, doesn’t reassure me that the outcome
would be the optimal.

And yes, the safety measures you list sound reasonable. However, we still
have much less control (in a sense of knowing absolutely what we are doing,
in terms of functions and consequences) when hacking genetic material
of living beings, than hacking with the source (or machine) code of computer programs.

It is in the latter field, that we _in principle_ have the total control of
what we are doing. I.e., we KNOW the purpose of each instruction and
subroutine working in 100% deterministic automaton.
Bearing this in mind, it would not be at all impossible to write 100% safe programs
with no security holes for “worms” or other computer parasites.
However, the status of our computer and network security is still dismal!

Now contrast that to a situation with the genetic manipulation,
that we are effectively just GUESSING the “purpose(s!)” of each gene, which are
acting in widely differing genomes (bifurcating from each other
as the evolution proceeds, which is quite fast with microbes).

Here are some examples of horizontal transfer with plasmids:

http://iai.asm.org/cgi/content/abstract/48/1/241

http://www.pnas.org/content/96/24/14043.full.pdf

That NPR story is a good find. Thanks.

Anyway, regarding GM microbes, this is what the GreenTech article wrote:

“The biologically produced BDO is chemically identical to regular BDO, but it costs far less to produce and far less energy gets consumed in the manufacturing process, he added. Making regular BDO involves cooking up several different chemical intermediates at high temperatures. Biological BDO requires genetically enhanced microbes (E. coli by the way) and an environment that allows them to grow and breed. The reaction takes place at normal air pressures and requires only slightly higher-than-normal room temperatures.”

If you are concerned with chemical contamination of our environment, I think this is a big breakthrough. Sure, BDO is toxic, but I’d bet that many of the intermediates used in its production are also toxic. When a microbe is used to produce this compound, manufacturers no longer need to keep large stocks of the intermediates, which inevitably will leak into the environment. Instead, the intermediates are produced and consumed with high efficiency within a single cell.

If you are concerned with a microbe producing a toxic compound, my only thought can be “so what?” Microbes already produce a wide variety of toxic compounds. That’s why we don’t eat rotten food.

Okay, so maybe the choice of E. coli (a gut commensal) is bothersome. But the chance of these GM E. coli colonizing our guts is virtually zero. First, engineers typically use laboratory strains that have been extensively modified from the wild-type that was isolated from a human host (50 years ago, if they are using the standard lab strain). Not only have these strains adapted to the laboratory environment over the past 50 years, but engineers typically make extensive changes to them in order to facilitate the engineering process. One strategy is to systematically knock out any gene that isn’t necessary in the laboratory environment; these bugs wouldn’t know what to do if they were introduced into an intestine and had to compete with wild E. coli. Another reason that these microbes would be at a competitive disadvantage in the gut is that they are designed to export large amounts of high-energy compounds (BDO)– wild competitors would not bear that same burden, and even for this strain, a single mutation could knock out the BDO production function and provide a huge boost in the energy available to the microbe..

http://www.npr.org/templates/story/story.php?storyId=102893816

Nice idea, doesn’t work. That’s the PIO problem I mentioned; putting people in a control loop like that actually makes instabilities worse. The thinking for the proposed solution is wrong, too. It’s what led to people thinking it would be safe to introduce cane toads to Australia deliberately as a biological control for a previous mistake, because they had learned from previous mistakes and were testing for dangers and applying safeguards. The pattern of introducing a flawed fix for a previous flawed introduction – on a human response time scale – is a case of PIO at work. And, of course, there is the scope for malice, both at the “mad scientist” level and at the level of introducing animals that were tailored for one environment to another, say by teenage boys deliberately catching a few and transporting them.

He then suggests “If we take this view, then we don’t need to reject GM agriculture overall…”

No! No! No! What counts isn’t whether you have a view, but whether it is right. The former only works for shifting blame, not for stopping harm.

Sorry, I have no formal published work on the differential equation instabilities, I merely commented on this area in private communications, in the light of that background knowledge of mine.

“Is your concern with gene flow from GM crops into wild populations, or with the behavior of the crops in the absence of wild relatives?” It’s to do with both, at an ecological level as well as species or individual level. Since we can actually see such things with introduced natural species, we know that at least as much problem will come up with genetic engineering – plus, the shortened time scale issues and the indefinite supply of new variation.

JohnElder writes “Other than a James Bond villain, I cannot imagine anybody setting out to create such horrors, and some may be impossible”. Curiously enough, I have just submitted a short story along those lines to a competition linked at Ken MacLeod’s blog (http://kenmacleod.blogspot.com), about someone devoting his life to finding a small change that would wipe out the human race. But it isn’t truly fiction, except in the details. Almost precisely the technique I used in the story is being pursued by John Abramyan, supervised by Professor Peter Koopman (http://www.invasiveanimals.com/about/our_students/john_abramyan/index2.html), looking for a way to deal with the aforementioned cane toads. The problem here is twofold: cane toads from Australia could be taken back to the original populations inadvertently or maliciously, so wiping out a chunk of a working ecology; and, the more the techniques get developed and applied, the lower the barrier gets for the same thing being done where we don’t want. Part of that lower barrier is just that the techniques get easier and cheaper, but part is like (say) financial regulatory systems; the longer things work, the more casual the regulators get.

or

http://greenlight.greentechmedia.com/2008/09/16/genomatica-gets-microbes-to-make-industrial-chemicals-560/

Note that the chemical in question, “One Four B-D-O”, has similar effects
as the “popular” drug “gamma”. Now, just go figure what kind of diseases
we might soon have.

More innovations from this shining new company:
http://www.greentechmedia.com/front2/sitesearch/results.jsp?si=1&sl=3069&searchterm=genomatica

Nanotechnology is not necessarily capital intensive; it’s open to the citizen biologist with modest resources.

http://diybio.org/