…in 10 years we’ll have the technology to construct human genomes – from scratch. …So what the genetic engineer of the future wants is a laptop computer hooked up to a database of standard bio-parts [see the Biobricks Foundation, and MIT's Registry of Standard Biological Parts]. On that computer I want to run software that allows me to mix and match the parts to define a big piece of DNA that will be the program for the bio-system I want to build. The design I develop with my program gets shipped to a DNA synthesis company which ships the finished DNA back to me by overnight express. Think of chip design/chip fab — but for living systems.
Yesterday, while cycling the Derwent River in Hobart, I listened to Futures in Biotech podcast 008 — an interview with MIT prof. Dr. Drew Endy titled Genetic Engineering 2.0: Synthetic Biology. The MP3 is here. Be warned that the audio quality on this episode is poor – but the content more than justifies the effort.
Dr. Endy [Wikipedia] is an engineer’s engineer — one of those who are building the reality predicted by Ray Kurzweil’s “Law of Accelerating Returns“. For those who think Kurzweil is over-the-top optimistic in his forecasts, I highly recommend the entire Futures in Biotech series. After you hear what the researchers in the field are accomplishing you’ll understand where Kurzweil is coming from. Endy’s overview of what his MIT lab is up to:
We are working to enable the design and construction of large scale integrated biological systems. Biology presents a new medium for engineering and contains many domain-specific challenges (e.g., evolution). Still, in getting started, we can make use of past successful experience in other disciplines. We are currently exploring the application of three past engineering lessons: (1) standardization of components, conditions, and characterization, (2) abstraction as a tool for hiding information and managing complexity, and (3) decoupling of overwhelming, complicated problems into many simpler problems (e.g., design and fabrication).
In this podcast Dr. Endy explains several key elements of synthetic biology. E.g., DNA synthesis, which is the interface between the design/modeling software and the wetware — effectively “a matter compiler for the stuff that programs living organisms”.
Paraphrasing Dr. Endy: So what the genetic engineer of the future wants is a laptop computer hooked up to a database of standard bio-parts [see the Biobricks Foundation, and MIT's Registry of Standard Biological Parts]. On that computer I want to run software that allows me to mix and match the parts to define a big piece of DNA that will be the program for the bio-system I want to build. The design I develop with my program gets shipped to a DNA synthesis company which ships the finished DNA back to me by overnight express. Think of chip design/chip fab — but for living systems.
Here’s how Endy arrives at the forecast of synthesizing a particular human genome in 10 years [there are a couple of problems in his math, but I'll accept the thrust of the argument]:
• a typical gene = 1000 base pairs
• 5 years ago it cost $10 to $16 per base pair, say $10,000 to synthesize a kilo-base-pair gene
• the cost has dropped by 2x each year of last 5 years, so now about $.79 per base pair
• note that DNA is really good at making copies of itself – for free, so once synthesized I can make copies for approx no money
• in 3 more years the cost will be about $.10 a base pair, so the gene will cost $100 instead of $10,000
• all the low level technologies to do this exist today – but are not yet integrated into an industrial scale process [note: Endy is the founder of a company, Kodon, whose mission is to do just that]
• today we have synthesized virus genomes
• by the end of 2007 we should have synthesized some much larger bacteria genomes
• “in 10 years we’ll have the technology to construct human genomes – from scratch”
• there are already projects proposed to rebuild baker’s yeast, 12 million base pairs, which at todays prices would cost about $8 million
• the human genome is 3 to 4 billion base pairs, so it would cost you $2 to 3 billion to synthesize today [less than the net worth of a number of individuals]
• “it would be shortsighted not to imagine that the technology will exist to make that accessible to many people in a decade”
One example Drew gave of what can be accomplished today with the available software, bio-parts libraries and DNA synthesis is the IGEM, or International Genetically Engineered Machine competition [described in this MIT Technology Review article]. One team designed and built a new type of e-Coli bacteria “without the stink” as described in this Boston Globe article
A group of MIT students, who wore black T-shirts emblazoned with their team name, “eau d’ecoli,” stood in the hallway asking passersby to sniff test tubes filled with E. coli bacteria and mark what each tube smelled like: banana, mint or “stinky.”
E. coli, a microbiology lab staple, smells like feet on a good day, one student said. Another compared it to “fecal matter.”
“Our sense of smell was totally obliterated by this nasty E. coli,” said Veena Venkatachalam, an MIT sophomore majoring in chemistry and physics. “We’re just trying to save all of us from this terrible fate.”
Via genetically engineered E. coli, the students managed to turn their lab into a more pleasant work environment that smells like a bakery. They engineered the E. coli to smell like mint while it was growing and to smell like banana when it was done.
Thus the banana introducing this post. On IGEM, Technology Review concluded:
As the number and complexity of parts grow, both students and industry and academic scientists can make ever-more-complicated designs. The machines entered in the 2006 iGEM competition have doubled in size in the past two years, from about 6,000 to 12,000 letters of DNA. “These [projects] represent the largest designed genetic systems that have ever been developed,” says Chris Voigt, a bioengineer at the University of California, San Francisco, who is advising one of the student teams. “Understanding how to push the size and complexity of these systems is what is going to have an impact.”
For another angle on Drew Endy’s work, try this podcast from the Open Source Conference of 2005 on IT Conversations. Endy quotes a “scientist” in the film Red Planet – the one part they got right:
…I’m a geneticist, I write code…
UPDATE: I’ve located an excellent video of Drew Endy’s talk at 2006 Second International Conference on Synthetic Biology “On a System for Engineering Genetic Machines”. Highly recommended.
Technorati Tags: Singularity
quite informative!