Radar Theme: Synthetic Biology

[This is part of a series of posts that briefly describe the trends that we’re currently tracking here at O’Reilly]

Drew Endy taught undergraduate students how to make e. coli bacteria that smelled like wintergreen, using his biobricks. This shows us a future for biology where “useful biological tasks” can be “automated” using “components”. The quotes indicate where research and development are going—building components, figuring out how biological amateurs can assemble them, and to what end. The overlap with open source and the low-barrier-to-entry that’s reminiscent of the web are particularly interesting to us.

Watch list: Drew Endy, George Church, Christina Smolke, Open Wetware, Ginkgo Bioworks.

tags: ,
  • This feels like the one area where the human race is trembling on the verge of opening Pandora’s box; if undergrads can gin up Christmassy e. coli, will graduate students be able to craft something that has the effects of Marburg, but with a longer latency and the contagious capability of the flu? I think we ought to give a lot more attention to making our world less susceptible to mass contagion and its effects, which include inceasing biodiversity (so one bio-tinkerer can’t take out the whole of the Kansas wheat crop), and making social structures more robust.

  • tom curtin

    I think it’s pretty easy to always have a doomsday scenario at hand, but there are many many more positive outcomes. Drew’s approach to me is the wiser one, building from the ground up the way programmers build from proven libraries of code.

    Most genomics work to me is like taking several dumps of Windows vista binary code (chosen randomly as an example of an OS with bugs) running on several different physical platforms and then combining, moving or flipping strings of bits experimentally until you find something interesting and then saying “AHA Look what I’ve invented! Give me some money now!”

    AT least the synthetic biologists don’t have to worry about “God” hitting them for code infringement, and their code is documented!

  • chuck brownstein

    Making new living stuff that survives long enough to mess with old living stuff indeed seems as likely as doing something executable with random Vista code chunks. I guess that may back Ross’s point.

  • chuck brownstein

    Making new living stuff that survives long enough to mess with old living stuff indeed seems as likely as doing something executable with random Vista code chunks. I guess that may back Ross’s point.

  • In response to Tom Curtin, above:

    I see two large problems with the way this research is conducted:

    1) The wet labs and the farmers are far too casual about containment. Feral descendants of synthbio toys are a virtual certainty. Nobody knows how they’ll evolve. We only know that they reach a feral state by end-running around natural checks and balances: humans spend lots of energy introducing them to the environment. We’re basically “shaking the rafters”.

    2) The interpretation of results, at least in popular understanding and in the vision Dr. Endy promotes is vastly over-reaching. The notion of a “component-based” system for genetic engineering is based on an assumption. The assumption is that, by nature, genetic form and function follows a modular design. You can’t have components without a modular architecture, by definition. That assumption is (demonstrably) true by some metrics and false by others. For example, phenotypical traits that follow patterns of Mendalian heritability are, in that regard, a modular feature — the assumption holds by that metric. On the other hand, metabolic function is a clear example where modularity is often violated, given the complex interactions between the expressions of many different genes in a single process — the modularity assumption is partially false in that case. There are unknown areas, too, and one that is particularly alarming: mutation rates and locations do not appear to be random, as far as I understand things. Yet, when we apply the “modularity assumption” and make e. coli smell like chewing gum (or whatever), we simply do not bother to measure how mutations play out in offspring of our creations (the self-same creations we handle with such casual attention to containment).

    Summing up the above: we don’t really deeply understand what we’re doing in synthbio and we are being sufficiently lazy to ensure that the biosphere as a whole will be the test tube. This does not seem prudent. Endy et al’s vision of a new kind of engineering is at best premature and, because of the problems with the “modularity assumption” may not prove out at all — yet whatever is accomplished in the attempt shall surely be released into the wild.