h+ Magazine

Winter 2009.

Issue link: http://cp.revolio.com/i/5039

Contents of this Issue


Page 49 of 89

ARTwORk COURTESy Of yASAMAn SHERI http://yeahsnos.wordpress.com/ http://yaz.boti.ca/ie/ 50 winter 2009 one of the ideas I champion is that Dna is a programming language for living things. By stringing Dna bases together in different ways, one gets different organisms. With one sequence, a bacterium is the result. With another, a butterfly. The same can be said about any subcomponent of life, all the way down to individual proteins. As we get better at "printing" DNA with automated synthesizers, it gets easier to make DNA-based programs, from simple scripts (instructing a bacterium to make a new protein or compound) to whole new operating systems (genomes). And it's just gotten easier, faster and cheaper — a biological version of Moore's law. With DNA synthesis, metabolism can be shaped by anyone who can master various DNA design tools. It's the start of a whole new era in biology: digital biology. I started focusing on DNA synthesis about ten years ago. At the time, I worked for a large biopharmaceutical company. As with any language, mastering DNA means one must learn to read, comprehend, and write. We had a fantastic bioinformatics team. We bought a subscription to Celera, the company Craig venter created to sequence the human genome. With reading and comprehension well taken care of, it made sense to start thinking about how to write DNA code better. Celera was possible because people had spent decades improving DNA sequencing technology. Still, the state of the art of DNA synthesis was poor, with low throughput and high cost (on the order of $10 per base pair). Making even a small protein (roughly 1000 bases) was expensive, and only justifiable for things like small, high-value proteins such as a growth hormone. But I believed that as synthesis costs fell over time, less lucrative applications or experimental designs that had a higher probability of failure would fall within reach. Moreover, the work would become increasingly computer- based, rather than being done in the laboratory. Genetic engineering would come to resemble software engineering, except the programming would be biochemical. In 2003, I took a year off to digest past experiences and to consider where life science may be going in the near future. In the meantime, digital biology got a name: synthetic biology. A small group at MIT was leading the way with DNA modules they called BioBricks that could be snapped together like lego blocks and then easily reconfigured. The next year, they developed a student training program with BioBricks and challenged student teams to be creative in designing and making applications. Almost overnight, the genetic engineering capability once available only to the experienced and well-financed became available to relative novices for a fraction of the price. Around this time, I found myself thinking a lot about open source versus proprietary software. The success of open source software, like linux and Apache server, had demonstrated that community-based development could rival the work done in dedicated companies. Was open source biology possible? I believed strongly that synthetic biology, done openly, could eventually compete with the for-profit biotechnology industry. I could see a day where almost anyone with a laptop could start to create software for cells. What would people make? The projects developed by students with BioBricks suggested a broad range, from fun (bacteria programmed to smell like bananas or wintergreen) to commercially useful (next generation biofuels like butanol). By 2005, several synthetic biology companies had appeared. They'd attracted large investments from top-tier venture groups. The field was hot. I began to think seriously about creating a linux-style company to make drugs. How would the company be financed? How would people work together? What would they make? Eventually, I came to believe that drug development needed a complete reboot. In the wake of the Human Genome Project and increasing lab automation, life science data was exploding. Genomics had spawned proteomics and metabolomics, and even more "omics" were appearing on the horizon. Research was growing exponentially, but development,was still stuck on a linear path from discovery to the clinic that could take a decade and a billion dollars or more. The gulf between biological R&D, always wide compared to more traditional fields of engineering, was growing even wider. Andrew Hessel of Pink Army Cooperative on Forming the First DIY Drug Company

Articles in this issue

Archives of this issue

view archives of h+ Magazine - Winter 2009.