I am sitting in the Stata Center at MIT taking a breather from serving as a judge at International Genetically Engineered Machines 2009 Jamboree. There are 110 teams here, with over 1200 students from around the world showing off their projects with great enthusiasm. As we have a full day left to go before the deliberations begin I won't divulge yet how specific teams are doing. But I have to say I am pleased.
iGEM is, at its core, an experiment. As the wiki says, the teams will "all specify, design, build, and test simple biological systems made from standard, interchangeable biological parts." Of course, as there aren't yet any standard, interchangeable biological parts, the students are inventing as they go. And inventing is slow, arduous work.
The most impressive talks I have seen this year do not represent giant leaps forward in new biological technologies (though some of the projects are real steps forward in that regard). Rather, I have been pleasantly surprised that many teams took up the challenge of improving or better characterizing parts that were already in the registry. Many of those parts don't work as advertized, or do not have enough data in the registry to know how they really work. That will slowly get fixed.
That it will take time to get all this working can make the differences between the annual Jamborees appear slight. Thin film semiconductors themselves took decades to get working, and even then those systems were built on top of a good century and a half of practical experience with electricity and then basic electronics. iGEM is attempting to squeeze all that effort into just a few years.
I am put in mind of W. Brian Arthur's work on the dependence of innovation on the availability of components. Here is a recent review of his book, The Nature of Technology. Historically, and theoretically, the complexity of technological artefacts tends to increase in leaps and bounds as components are combined in new ways, and then combinations then serve as components for the next generation of innovation. But first you have to have functioning components.
Drew Endy asked me yesterday if I thought we were stuck in a rut. Nope. Just stuck in reality.
This week Biodesic shipped an engineering prototype of the LavaAmp PCR thermocycler to Gahaga Biosciences. Joseph Jackson and Guido Nunez-Mujica will be showing it off on a road trip through California this week, starting this weekend at BilPil. The intended initial customers are hobbyists and schools. The price point for new LavaAmps should be well underneath the several thousand dollars charged for educational thermocyclers that use heater blocks powered by peltier chips.
The LavaAmp is based on the convective PCR thermocycler demonstrated by Agrawal et al, which has been licensed from Texas A&M University to Gahaga. Under contract from Gahaga, Biodesic reduced the material costs and power consumption of the device. We started by switching from the aluminum block heaters in the original device (expensive) to thin film heaters printed on plastic. A photo of the engineering prototype is below (inset shows a cell phone for scale). PCR reagents, as in the original demonstration, are contained in a PFTE loop slid over the heater core. Only one loop is shown for demonstration purposes, though clearly the capacity is much larger.
The existing prototype has three independently controllable heating
zones that can reach 100C. The device can be powered either by a USB
connection or an AC adapter (or batteries, if desired). The USB connection is primarily used for
power, but is also used to program the temperature setpoints for each
zone. The design is intended to accommodate additional measurement capability such as real-time fluorescence monitoring.
We searched hard for the right materials to form the heaters and
thin film conductive inks are a definite win. They heat very quickly
and have almost zero thermal mass. The prototype, for example, uses
approximately 2W whereas the battery-operated device in the original publication
used around 6W.
What we have produced is an engineering prototype to demonstrate materials and controls -- the form factor will certainly be different in production. It may look something like a soda can, though I think we could probably fit the whole thing inside a 100ml centrifuge tube.
The prototype necessarily looks a bit rough around the edges as some parts were worked by hand where they would normally be done by machine (I never have liked working with polycarbonate). We have worked hard to make sure that the LavaAmp can be transitioned relatively seamlessly from prototype quantities, to small lot productions, to high-volume production. The electronic hardware is designed to easily transition to fabrication as a single IC, all the plastic bits can be injection molded, and the heater core can be printed using a variety of high-throughput electronicss fabrication methods.
Next up will be field trials with a selected group of labs, as well as more work on refining the loading of the loops.
A fortnight ago the World Wildlife Fund released a report pushing industrial biotech as a way to increase efficiency and reduce carbon emissions. Interesting. Of course, industrial biotech doesn't necessarily require direct genetic modification, but the WWF must know that is an inevitable consequence of heading down this road. More on this after I get a chance to read the report.
News today that a federal judge has rejected the approval of GM sugar beets by the USDA. The ruling stated that the government should have done an environmental impact statement, and is similar to a ruling two years ago that led to halting the planting of GM alfalfa. As in that case, according to the New York Times, "the plaintiffs in the [sugar beet] lawsuit said they would press to ban planting of the biotech beets, arguing that Judge White's decision effectively revoked their approval and made them illegal to grow outside of field trials." The concern voiced by the plaintiffs, and recognized by the judge, is that pollen from the GM beets might spread transgenes that contaminate GM-free beets.
A few other tidbits from the article: sugar beets now supply about half the US sugar demand, and it seems that GM sugar beets account for about 95% of the US crop (I cannot find any data on the USDA site to support the latter claim). A spokesman for the nation's largest sugar beet processor claims that food companies, and consumers, have completely accepted sugar from the modified beets -- as they should, because it's the same old sugar molecule.
I got lured into spending most of my day on this because I noticed that the Sierra Club was one of the plaintiffs. This surprised me, because the Sierra Club is less of a noisemaker on biotech crops than some of the co-plaintiffs, and usually focuses more on climate issues. Though there is as yet no press release, digging around the Sierra Club site suggests that the organization wants all GM crops to be tested and evaluated with an impact statement before approval. But my surprise also comes in part because the best review I can find of GM crops suggests that their growing use is coincident with a substantial reduction in soil loss, carbon emissions, energy use, water use, and overall climate impact -- precisely the sort of technological improvement you might expect the Sierra Club to support. The reductions in environmental impact -- which range from 20% to 70%, depending on the crop -- come from "From Field to Market" (PDF) published earlier this year by the Keystone Alliance, a diverse collection of environmental groups and companies. Recall that according to USDA data GM crops now account for about 90% of cotton, soy, and corn. While the Keystone report does not directly attribute the reduction in climate impacts to genetic modification, a VP at Monsanto recently made the connection explicit (PDF of Kevin Eblen's slides at the 2009 International Farm Management Congress). Here is some additional reporting/commentary.
So I find myself being pulled into exploring the cost/benefit analysis of biotech crops sooner than I had wanted. I dealt with this issue in Biology is Technology by punting in the afterword:
The broader
message in this book is that biological technologies are beginning to change
both our economy and our interaction with nature in new ways.The global acreage of genetically
modified (GM) crops continues to grow at a very steady rate, and those crops
are put to new uses in the economy every day.One critical question I avoided in the discussion of these
crops is the extent to which GM provides an advantage over unmodified
plants.With more than ten years
of field and market experience with these crops in Asia and North and South
America, the answer would appear to be yes.Farmers who have the choice to plant GM crops often do so,
and presumably they make that choice because it provides them a benefit.But public debate remains highly
polarized.The Union of Concerned
Scientists recently released a review of published studies of GM crop yields in
which the author claimed to "debunk" the idea that genetic modification will "play
a significant role in increasing food production"The Biotechnology Industry Organization
responded with a press release claiming to "debunk" the original
debunking.The debate continues.
Obviously we will all be talking about biotech crops for years to come. I don't see how we are going to address the combination of 1) the need for more biomass for fuel and materials, 2) the mandatory increase in crop yields necessary to feed human populations, and 3) the need to reduce our climatic impacts, without deploying biotech crops at even larger scales than we have so far. But I am also very aware that nobody, but nobody, truly understands how a GM organism will behave when released into the wild.
Last week's Economist has another story on biohacking, "Hacking goes squishy", that contains a nod or two to the economic context and also has a version of my cost curves. I have a couple of thoughts.
The cost curve figure in the article was finished early August, and since then I have decided to add additional data points. Just a couple of days ago, an ad from Mr. Gene (a division of GENEART) showed up in my inbox advertising synthesis for $.39/base pair. I haven't had time to figure out why GENEART itself charges $.44/base, but presumably there is some additional customer service/sequencing/etc. thrown in. The latest commercial cost for oligos (in low volume) appears to be about $.15/base, which is actually a slight increase compared to prices I found a couple of years ago. More on this later.
On the sequencing side of things, Illumina has delivered its first commercial human genome at $48,000. Here is the Bio-IT World summary: "Illumina completed the sequence at its CLIA-certified laboratory, producing more than 110 billion base calls, good for 30X coverage of the genome and the identification of some 300,000 novel single nucleotide polymorphisms." I'll call it $8x10^(-6) per finished base, even though they actually sequenced many more bases than are in a human genome.
In other sequencing news, Complete Genomics just announced (PDF) the sequencing of 14 individuals for various academic projects. They claim to be on track to offer $5000 human genomes in the next six months. Helicos made a lot of noise last month with the publication of Steve Quake's genome at a cost in reagents of ~$48,000. While all the numbers in the article are impressive, like many observers I still have questions about the actual cost per base in a commercial operation. Labor? Cost of capital? Nonetheless, the technology is impressive.
The cost of sequencing continues to fall rapidly. The race to the bottom is well under way. Here is the figure:
It is interesting that the oligo and gene synthesis numbers have the appearance of slowing down. I don't believe this is evidence of a real trend, but rather that the cost of synthesis is now about labor and finance rather than about raw materials. And sequencing (proof reading) of synthetic genes now accounts for a good hunk of the cost, depending on what exactly you are synthesizing. Since I have now seen several different technologies that can be used to reduce costs, I expect prices to continue falling in the years to come. One technology nearing the marketing stage enables the use of unpurified oligos in gene assembly, including those synthesized on chips, through true error correction rather than error removal. While the consequent reduction in the cost of raw materials may not add up to much, there should be substantial cost improvements from 1) reducing required sequencing and 2) the ability to automate assembly.
I can also now update my "genetically modified domestic product" (GMDP) numbers for the US. My earlier article "Laying the foundations for a bio-economy" (journal link), contained an estimate that genetically modified systems generate revenues that are the equivalent of about 1% of US GDP. It turns out that is too small.
The reason for the underestimate is that I was overly trusting of reporting by The Financial Times, Nature Biotechnology (upper left panel), and others, who all published stories claiming that 2007/8 revenues from genetically modified crops were about US$ 8 billion worldwide and a bit over $4 billion in the US. It is interesting to me that all these organizations misreported in exactly the same way a number published by the ISAAA in its report "Global Status of Commercialized Biotech/GM Crops: 2008". In their defense, the reporters probably just had access to the executive summary, which contains the phrase "the global market value of biotech crops... was US$7.5 billion", and they were probably in a hurry to meet deadlines. But the very next sentence in the executive summary reads "The value of the global biotech crop market is based on the sale price of biotech seed plus any technology fees that apply." So that ~$8 billion worldwide is just seeds and related fees. And seeds grow. Into bigger things. With greater value. Like crops.
A quick visit to the USDA reveals US revenues from GM crops that is in the neighborhood of $100 billion. Here is a nice figure showing crop adoption since 1996, which gives us the percentage of acres planted in GM seeds. Then, jumping over to the National Agricultural Statistics Service, you can figure out the revenues per crop. Put it all together and you find out that in 2007 the value to US farmers of revenues from GM crops was close to $70 billion.
Here is a table from Biology is Technology that lays out some of the global numbers up through 2007:
Table 11.1
Revenues from genetically modified systems in 2007
Sector
Worldwide revenues
($ billions)
US revenues
($ billions)
% of US GDP (total of $~14
trillion)
Revenue growth rate in US (%)
Biotech drugs ("biologics")
79
67
.48
15-20
Agbiotech/GMOs
128 (est)
69
.49
10
Industrial
~110
~85
.61
15-20
I left out of the book any discussion of what benefit GM crops give compared to non-GM crops because I don't yet trust any of the numbers I have found; estimates range from -30% to +30%. When I have time to sort it out for myself, I will publish something. Until then, I would note that it seems unlikely to me that farmers around the world would keep buying GM seeds (that are more expensive than non-GM seeds) -- and buying more GM seeds every year -- if they didn't benefit financially from making that choice.
By the way, for those who have asked or are curious, I just learned that the book comes off the presses in the first week of December, though I don't know when they actually will be available in stores and whatnot. No news yet on e-versions for the Kindle, etc., but let me know if you are interested.
Anyway, although not all the numbers for 2008-2009 are available (including GDP), at this point I am pretty comfortable with the estimate that revenues from GM systems in 2009 will be the equivalent of about 2% of US GDP. That is a big number. As big as mining in the US. And there is no way mining is growing at ~15% a year. The future of the economy is biology.
Biodesic evaluated systems biology investments for a large organization about 18 months ago, and Schadt's approach makes more sense to me -- by far -- than anything else we looked at. I sat in on the pitch that Schadt and Stephen Friend made to that sameorganization, and it was crystal clear to me that Sage -- now residing
at the Hutch here in Seattle -- should be on the receiving end of piles of money. The stacks of Nature Group publications Schadt is accumulating suggest he is on to something, and it appears that his methods can be used to make predictions about the behaviors of complex networks. Time and experimentation will tell, of course. The open source aspect is a huge bonus.
Schadt's move to Pacific Biosciences is interesting because during his talk he suggested that genome sequencing provides enough information about variation to fuel his statistical methods for predicting interactions not just between genes but between tissues -- he is working at the level of describing the behavior of networks of networks. It seems he will now have access to plenty of data.
My first published effort at tracking the pace and proliferation of biological technologies (PDF) was published in 2003. In that paper, I started following the efforts of the DEA and the DOJ to restrict production and use of methamphetamine, and also started following the response to those efforts as an example of proliferation and innovation driven by proscription.
The story started circa 2002 with 95% of meth production in Mom and Pop operations that made less than 5 kg per year. Then the US Government decided to restrict access to the precursor chemicals and also to crack down on domestic production. As I described in 2008, these enforcement actions did sharply reduce the number of "clandestine laboratory incidents" in the US, but those actions also resulted in a proliferation of production across the US border, and a consequently greater flow of drugs across the border. Domestic consumption continued to increase. The DEA acknowledged that its efforts contributed to the development of a drug production and distribution infrastructure that is, "[M]ore difficult for local law enforcement agencies to identify, investigate, and dismantle because[it is] typically much more organized and experienced than local independent producers and distributors." The meth market thus became both bigger and blacker.
Now it turns out that the production infrastructure for meth has been reduced to a 2-liter soda bottle. As reported by the AP in the last few days, "The do-it-yourself method creates just enough meth for a few hits, allowing users to make their own doses instead of buying mass-produced drugs from a dealer." The AP reporters found that meth-related busts are on the increase in 2/3 of the states examined. So we are back to distributed meth production -- using methods that are even harder to track and crack than bathtub labs -- thanks to innovation driven by attempts to restrict/regulate/proscribe access to a technology.
And in Other News...3D Printers for All
Priya Ganapati recently covered the latest in 3D printing for Wired. The Makerbot looks to cost about a grand, depending on what you order, and how much of it you build yourself. It prints all sorts of interesting plastics. According to the wiki, the "plastruder" print head accepts 3mm plastic filament, so presumably the smallest voxel is 3mm on a side. Alas this is quite macroscopic, but even if I can't yet print microfluidic components I can imagine all sorts of other interesting applications. The Makerbot is related to the Reprap, which can now (mostly) print itself. Combine the two, and you can print a pretty impressive -- and always growing -- list of plastic and metal objects (see the Thingiverse and the Reprap Object Library).
How does 3D printing tie into drug proscription? Oh, just tangentially, I suppose. I make more of this in the book. More power to create in more creative people's hands. Good luck trying to ban anything in the future.
Various hard drive crashes have several times wiped out my records for the longest published synthetic DNA (sDNA). I find that I once again need the list of references to finish off the edits for the book. I will post them in the open here so that I, and everyone else, will always have access to them.
1979 Total synthesis of a gene HG Khorana Science 16 February 1979: Vol. 203. no. 4381, pp. 614 - 625 http://www.sciencemag.org/cgi/content/abstract/203/4381/614
1990 A totally synthetic plasmid for general cloning, gene expression and mutagenesis in Escherichia coli Wlodek Mandecki, Mark A. Hayden, Mary Ann Shallcross and Elizabeth Stotland Gene Volume 94, Issue 1, 28 September 1990, Pages 103-107 http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T39-47GH99S-1J&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=84ca7779ff1489d5e18082b9ecb80683
1995 Single-step assembly of a gene and entire plasmid from large numbers of oligodeoxyribonucleotides Willem P. C. Stemmer, Andreas Crameria, Kim D. Hab, Thomas M. Brennanb and Herbert L. Heynekerb Gene Volume 164, Issue 1, 16 October 1995, Pages 49-53 http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T39-3Y6HK7G-66&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=83620e335899881aac712a720396b8f2
2002 Chemical Synthesis of Poliovirus cDNA: Generation of Infectious Virus in the Absence of Natural Template Jeronimo Cello, Aniko V. Paul, Eckard Wimmer Science 9 August 2002: Vol. 297. no. 5583, pp. 1016 - 1018 http://www.sciencemag.org/cgi/content/abstract/1072266
2004 Accurate multiplex gene synthesis from programmable DNA microchips Jingdong Tian, Hui Gong, Nijing Sheng, Xiaochuan Zhou, Erdogan Gulari, Xiaolian Gao & George Church Nature 432, 1050-1054 (23 December 2004) http://www.nature.com/nature/journal/v432/n7020/full/nature03151.html
Total synthesis of long DNA sequences: Synthesis of a contiguous 32-kb polyketide synthase gene cluster Sarah J. Kodumal, Kedar G. Patel, Ralph Reid, Hugo G. Menzella, Mark Welch, and Daniel V. Santi PNAS November 2, 2004 vol. 101 no. 44 15573-15578 http://www.pnas.org/content/101/44/15573.abstract
2008 Complete Chemical Synthesis, Assembly, and Cloning of a Mycoplasma genitalium Genome Daniel G. Gibson, Gwynedd A. Benders, Cynthia Andrews-Pfannkoch, Evgeniya A. Denisova, Holly Baden-Tillson, Jayshree Zaveri, Timothy B. Stockwell, Anushka Brownley, David W. Thomas, Mikkel A. Algire, Chuck Merryman, Lei Young, Vladimir N. Noskov, John I. Glass, J. Craig Venter, Clyde A. Hutchison, III, Hamilton O. Smith Science 29 February 2008: Vol. 319. no. 5867, pp. 1215 - 1220 http://www.sciencemag.org/cgi/content/abstract/1151721
The June 5 issue of Cell Stem Cells has a brief report describing the use of four proteins to reprogram human fibroblasts into induced pluripotent stem cells (iPSCs). I think this is a pretty important paper, as it dispenses with any sort of genetic manipulation of the target cells or any use of plasmids to insert new "control circuitry", or any chemical manipulation whatsoever.
As expected, it is getting easier to produce iPSCs, and the authors of the paper ("Generation of Human Induced Pluripotent Stem Cells by Direct Delivery of Reprogramming Proteins") note that their work demonstrates the elimination of "the potential risks associated with the use of viruses, DNA transfection, and potentially harmful chemicals and in the future could potentially provide a safe source of patient-specific cells for regenerative medicine".
Kim et al used four recombinant human proteins to turn human newborn fibroblast cells (purchased from ATCC -- see the Supplemental Data) into iPSCs, where each of the proteins was fused to a nine amino acid long "cell-penetrating peptide" (CPP) that facilitated the importation of the proteins across the cell membrane. The procedure was not particularly efficient, but after multiple treatments the authors produced cells that could differentiate into many different kinds of human tissues.
Here are a couple of thoughts about the paper. Note that in what follows I have only had a few sips of my first cup of coffee today, and my brain is still quite fuzzy, but I think I am mostly coherent. You can be the judge.
First, the authors did not use mature cells from adults, so don't expect this paper to lead to replacement organs and tissues tomorrow. The use of cells from newborns makes a great deal of sense for a first go at getting protein-based reprogramming to work, as those cells have already been demonstrated to be relatively easy to reprogram. The published procedure required many weeks of effort to produce iPSCs, and authors note that they have quite a ways to go before they can produce stem cells at the same efficiency as other techniques.
Nonetheless, it works.
Second, the paper describes PCR-based cloning of human genes to add the CPP sequences, along with a fair amount of bench manipulation to generate cells that made each of the four reprogramming proteins. All the sequences for those proteins are online, as are the sequences for the CPPs, so generating the corresponding genes by synthesis rather than cloning would now cost less than $10K, with delivery in 2-4 weeks. In another year, it will probably cost no more than $5K. (How long will it be before these proteins show up in the Registry of Standard Biology Parts?)
Third, the authors did not use purified reprogramming proteins to generate iPSCs, but rather used whole cell extracts from cells that produced those proteins. Thus the concentrations of the reprogramming proteins were limited to whatever was in the cell extract. This might critically affect the efficiency of the reprogramming. Presumably, the authors are already working on generating cultured cell lines to produced the reprogramming proteins in larger quantities. But if you wanted to do it yourself, it looks like you might "simply" have to order the appropriate sequences from Blue Heron already cloned into the human expression plasmid pCDNA3.1/myc-His A, which is available from Invitrogen. This would add a couple of hundred dollars to the cost because Blue Heron would have to play around with a proprietary plasmid instead of the public domain plasmids they usually use to ship genes. You would then follow the recipe from the Supplementary Data to transform a protein production cell line to make those proteins. Or perhaps you have a favorite recipe of your own. Here is something I don't get -- it looks like that particular expression plasmid adds a His tag to the end of the gene, so I don't understand why Kim et al didn't try a purification step, but maybe that is underway.
Fourth, if you wanted to do this at home, you could. You should expect to fail many times. And then you should expect to fail some more. And then, assuming your human cell culture technique is up to snuff, you should expect to eventually succeed. You might want to wait until the inevitable paper showing how to do this with adult differentiated skin cells is published.
And then what?
You will have an autologous stem cell line that you can use to produce tissues that are, immunologically speaking, identical to those in your body. What should you do with them? I would suggest you show them off at cocktail parties, brag about them on Facebook, and then destroy them with bleach and an autoclave. In lieu of an autoclave a microwave would probably do just fine.
But I expect that at least some of you will try to follow a recipe to generate some sort of human tissue, or even to simply inject those cells in your own bodies, which will result in all kinds of crazy teratomas and other tumors. To quote Harold Ramus, "that would be bad". So don't do that. Just because DIYStemCells are cool doesn't mean you should actually use them yourself. But I know some of you will anyway. That is the future of biological technologies, for better or worse.
