Joining 3.5 Billion Years of Microbial Invention

Good evening. (Applause) Thank you for coming I'm Stewart Brand from the Long Now Foundation I should say that most of you have these question cards and in this semi-darkness of the theater. It’s helpful when you’ve written out a question to come up to the front for Kevin Kelly and me to look at too. Later address the speaker with it’s helpful if you wave it. So, that the guys in the yellow hats can see it come collect it and bring it up. And you know write questions anytime during the talk, during the Q&A that’s the interactive part. Now, the Long Now Foundation is called the long now because we have in mind the next 10,000 years in the last 10,000 years and Peter Schwartz here is the one who came up with that number because 10,000 years ago there was a biotech revolution having to do with mainly plants and then animals that we eat and use. Now, we started genetic modifications, genetic engineering in a big way and humanity took a swerve that it’s still getting used to. And in about 150 years, another biotech revolution happened with biomedicine and humans started to be able to control their health, be able to control their birth rate and we get to the 21st century which regarded by many as the century of biology, which to a large extent means the century of biotechnology. And so these revolutions that we’ve seen seem to be on a kind of biotech Moore’s Law of getting more and more frequent and happening now in one lifetime instead of it millennia at a time. In the current acceleration of genetic science and genetic engineering is as often the case being carried by very sharp individuals. Scientist who find the right problems, find the right people to work with, find the right funding and swarm ahead. And really the exemplary case of that these days is Craig Venter. Please welcome him. Well thank you Stewart it’s certainly nice to be here with The Long Now Foundation taking a long view hopefully forward with Humankind. I think it’s very much an open question, but we’re trying to see if we can change some of the equations maybe to help that process along. Again, talking about two phases of information tonight, the first phase is gathering information. Actually referred to it and in my case is reading the genetic code and then we’ll turn to how now we’re using that information to change evolution, to change biology, to change hopefully some parts of society with writing the genetic code. And I’ll talk about the early stuffs of that. All this is happening I a pretty short period of time. We’ve sequenced the first genome of a living species Haemophilus influenzae in 1995. Before them we developed techniques for scaling up gene discovery. But we’ve only had complete genomes of living organisms for a relatively short period of time. It’s changing our view of looking at biology. We did two genomes in 1995 and I‘ll about the second one in the later phases of this had very different size genomes, very different characteristics. But after those first two that we had to fund ourselves we started getting a lot of government funding to scale up the process. We got funding to do most human pathogen. We’ve moved onto plants than simple animals. Then more complex ones such the fruit flies and people. The big change in technology after 1995 came surprisingly only four to five years later with a fruit fly genome. Haemophilus was 1.8 million letters of genetic code, Drosophila 180 million. And they were both done in the same four months time period. And 9 months later we had the three Billion first draft of the human genome. Computational aspects we’re very important. We actually had to build the third largest computer in the world to do this calculation. Now it’s not hard to do with much more standard computers. One and half tera-ops in 1999 was a huge computer. Today, we have things 10, 20, 30 and 100 times of that size. Just hits last year we published the first complete human genome. This was the diploid genome, sort of had both sets of chromosomes from both parents. What was done in 2000 and 2001 ended only being half the job. The public effort only got it half right because they set pout only to do half. Taking the other half would e easily discernible. The project of Solera actually tried to do the complete genome but we the mistake of overreaching we tried to do the genome of five individuals. There were three men and two women. No there was three women and two men. It sounds like a San Francisco movie, but because of the diversity, when we assembled the other genome. We ended up subtracting out all the major variation between them show the message and many you will remember it. In 2000-2001 as we all have the same set of genes and we all differ from each other only in one out of a thousand letters of genetic code. And this started all kinds of people down certain tracks. We have a pretty large industry now just to measure those one of thousand differences but it turns out we’re all 1 to 3 percent different from each other. So it’s a much greater variation than we imagined. To put that in context, we thought we differed from chimpanzees only by 1.27%. So you’re probably hoping that number changed as well or we really have explaining to do about evolution. It turns out we’re 4-5% different from chimps. And so this individual variation was totally missed. The two sets of chromosomes that each of you have. Each of those sets of chromosomes they don’t naturally have the same gene that the other pair has. For example there’s a major gene associated with detoxifying environmental toxins that of the Caucasian population a third have no copied of this gene. Another third have open copy and another third have two copies. But even when you have two copies there’s major variation of these. So it’s hard to find two of us with the exact same genetic code and we have a lot of complexity. I think it’s important that we know it now because it means we can move towards getting the right answers. In this first genome almost half the genes had a major variant in them and this certainly going to confound any body who was hoping to get simple interpretation of the human genetic code. So we have all kinds of companies promising that now for a thousand dollars. But it’s only a partial peek at what might be there. Seventy five percent of the letters the genetic code there variant between in any two of us. Didn’t show in the single nucleotide polymorphisms at that people were measuring. So, were scaling it out now to do 10,000 human genomes over the next decade. Hopefully, it won't take 10,000 years to it. But maybe only 10 and we will finally have the first chance to understand what’s genetic, what’s nature and what's nurture. Not only with the genetic code but were trying to collect intense extensive phenotypic information on all these individuals that were going to be sequencing and that’s a much greater challenge when you think about all the information that might describe you or your lives. How you look, how you think, your body structure, you organ structure, your metabolism all those things differ dramatically with each of us and trying to describe that makes sequencing the genome look like the easy part. We published our data in the public access of journal biology. You can download this 300 megabyte file, if you want to. It’s a zoomable version of the human genome that shows the types and extent of human variation and hopefully in a short time you’ll be able to get the map of your genome this way. Working with the X PRIZE Foundation, we now have a 10-million-dollar price for who ever gets the technology going to the extent that we can sequence an individual genome for a thousand dollars in a very short period of time. So, we’re hopeful that we’ll be giving out this 10 million dollars in the next 5 years, if you’re looking for a little spare change. You might try inventing something worthwhile. We’re releasing a new browser for the genome that allows it to be looked at down to the sequence file level as people for example get data from 23 in media or other places. They can look at actually a complete genome to try and put it some kind of context and find out how much is missing from those early test. Now, when we finish sequencing the human genome many people were hoping I would retire, but instead we looked around for what we thought were the most important projects in science to do particularly with that the technology that we developed for reading the genetic code to try and tackle. And it was clear to us that trying to look at the environment not only the extensive environment around us that helped influence our genes. But the broader environment to see if we could use DNA sequence as a new tool to see who we are sharing the planet with. There’s been this statistic for a long time that each milliliter of seawater has a million bacteria and 10 million viruses. But nobody really knew or what that meant and whether there was any diversity in that. And so we decided to do an experiment in the Sargasso Sea just taking a barrel of seawater filtering out all the organisms in it. Isolating the DNA from them on mass and sequencing it and just from one barrel of water we stopped sequencing after we had close to 1.4 million new genes. Maybe as 40,000 new species that had never been seen before. So the technique obviously worked because in fact people thought we’d find little or no life in the Sargasso Sea because it was supposed to be a desert with no nutrients in it. Again, I’ll show you why and how these organisms survived in a minute, but we decided to look further and we started the Sorcerer II expedition. I had always been looking for an excuse to sail around the world on my own vessel. And so we decided we would follow great scientific expeditions such as the Challenger expedition and sail around the world and taking samples every 200 miles. And sequencing everything we could find to see if we could generate a different view of life. And we were absolutely stunned with what we found and that was a special issue of PLoS that was published last year. This is the route that we followed, as with the Challenger expedition, we started in Halifax, went done the Eastern seaboard into the Caribbean Sea then down between Mexico and Cuba to the Panama Canal. Through the Panama Canal and down first Cocos Island and then to the Galapagos and that’s the dot you see here in blue. That’s what was covered in the special issue of PLOS of biology and that covered over 6 million new genes more than doubling the number of genes known from all science up to that point. And we’ve now been analyzing the samples from the rest of the globe After Galapagos, we left and sailed to the Marquesas where we are promptly arrested because there was a debate between the French government and the French Polynesia government as who had rights over the microbes that are there. You may not understand this but this is a physical map if you look at a political map in the Caribbean Sea for example there's no international waters. Every drop of water is claimed by one or more countries and you need permission from them to take a scientific example. Its okay to fish on most of those places you can shoot marine mammals you could do whatever you want. But if you’re asking scientific questions it’s considered extremely dangerous and you can get arrested for it. In fact, when the organisms in the ocean are out in the middle of the Pacific Ocean, they’re in international waters and they belong to nobody or to everybody. As soon as the one-knot current that goes across the Pacific carries them into French Polynesia all those microbes instantly becomes French genetic heritage. (Laughter) And they're willing to defend it to the extreme. We finally got out of there and took samples and spent some time in Australia then went across the Indian Ocean. We got halfway across and we and we stopped in Chagos Island, which is near where our country has a B-52 Air Base and we’re arrested again by the British this time because we would take samples and understand science in that water. So, science is something and organisms that are greatly protected around the world. And to be able to publish this date in the public databases we had to put geographic GPS coordinates on every DNA sequence. So, for example is somebody makes a discovery using the sequences that we found in Australia and you want to commercialize those you have to contact the Australian government and negotiate something that’s not clear what. So science is much more complicated today, but the experiments were very simple. We just simply filtered seawater through different size filters. We could then just take those filters put them in a freezer. When we got into a port, we put them in a FedEx package with some dry ice and sent them back to the lab in Rockford, Maryland where all the DNA was isolated at once from and sequenced and then reconstructed in the computer. And when we reconstructed it, we are amazed with what we found. For example, every 200 miles, 85% of the sequences and the organisms are unique. The ocean is not homogenous mixture. Samples that we’ve taken off of San Francisco Bay will be very different than the ones off of Los Angeles and off Seattle and Oregon. In fact if you look at the – so they’re red or dark here if you color blind is warm water the blue light colors is cold water. And even the 3% of sequences that assemble across more than one site. They’re an absolute distinction between a warm in cold water. And some of these change on an annual basis with weather. Other things that affect it or sunlight, But – and nutrients we can tell simply from a sample of seawater. Looking in the DNA in that water, where that water came from in the world? And for example, with all the ships that had come and gone from San Francisco Bay. If a tanker comes in and offloads it cargo, it fills up its hold with seawater. And we’ll go to its next port, dump all that seawater and take on new cargo. So, look at all the ships that come and go from San Francisco Bay each day and imagine with a million bacteria and 10 million viruses per milliliter. How these environments are constantly being challenged and shifted. One of the biggest discoveries we’ve made early on was that these organisms we discovered in the ocean had photoreceptors. Molecules very to our own visual pigments and almost everyone in the top parts of the ocean have these photoreceptors. The blue segment at the bottom of the slide is our prior knowledge of photoreceptors before this expedition. And these are very deep branching. A lot of diversities some set it would be in a rare gene family just affecting our visual acuity. It affects much life on the planet. We can line all these proteins; this was early on when we only had a couple of thousand of them. And why do this? It turns out there’s a single amino acid residue that determines the wave length of light that these receptors see. So then that allows us to ask some unique questions. Do we see any association with different geographic regions with the wavelength of light? And we’re quite surprised in fact to see something that makes sense. But nobody predicted ahead of time. For example, the Sargasso Sea, it’s a deep indigo blue the organisms that are there the photoreceptors see blue light. You get into a coastal waters where there’s a lot of chlorophyl, they see primarily green light. Yet in the fresh water like the Panama Canal they see entirely green light. So, there’s only one letter that genetic code then this needed to change to change the wavelength of light by changing this amino acid residue. And even though there is a large abundance of organisms traversing these waters. The ones that survive and grow are the ones that can get their energy directly from the sun. And last year there was a study from the Swedish group showing in fact these organisms grow directly on sunlight. This is not photosynthesis. This is the biological mechanisms that you are seeing in the slide with right now. You're having light hit these receptors and they transport ions across the membrane well in these microorganisms. That’s how they generate their energy in the low nutrient environment. So, in fact, instead, of low nutrients indicating and the scarcity of life this was the some of the highest density of life anywhere that we’ve seen because they have these mechanisms of getting of getting energy straight from sunlight The other thing that we’ve found is where people thought there was a single organism, we found evidence for thousands. All related in the sense they have similar sets of genes, similar gene order, but hardly a single one. Each one of these little bars you see on here represents 900 base pairs of genetic code. One of the things we can do is take a slice out of this date anywhere and create trees to look at the relationships between these very closely related organisms. They’re color coded by sights. So we can look at things like Atlantic Ocean versus the Pacific Ocean. But the bottom to me is the most exciting when you look at recent evolutions. There’s been a switch between blue and green light for these photoreceptors for different times So you can see at such selective pressure, you get a mutation and you switch from blue to green and you're in coastal waters obviously you're going to thrive. If you switch fro blue to green out in the middle of the ocean you probably won't. You can ask basically any question of this data in a similar fashion. There’s in fact a lot questions. How much novelty is there with all these new discoveries that have been doubling the number of all genes known to science. Are they like the photoreceptors just new members of known gene families? Or are we really making new discoveries and if we are making new discoveries what’s the pace of these discoveries? When we look at the data’s had initially we’re quite shocked to find that our database just between Halifax and the Galapagos was twice the size of all the public databases in terms of gene content. And so we did a calculation by comparing all those data to each other. It was about a million CPU hours and we found out thus far in the data set, there was maybe 50,000 major gene families. Well, we only have 22 – 23,000 genes, maybe, 14 or 15 gene families so even all of human biology that represents only a subset of this data but if we look at this as a cur instead of a long now, we have a long tale that basically goes out approaching infinity right now with new discoveries and new gene families with the exception of the animal world. So just to remind people, we’re just in one part of evolution even though the tree model doesn't really hold; it is useful for these kinds of diagrams. When we look at our reading the genetic code in mammals, it is basically been saturated so sequencing another Mammalian Genome, your favorite pet, the species down the street, whatever, you’re not going to discover any new genes. You might discover some unique combinations or spawning of those genes but we pretty much saturated the mammalian part of the tree but if we look at bacteria and archaea we’re in a linear phase of discovery. We don’t really know where we are, we just know it’s linear. Our thinking is that we’re in the earliest phases of this, just a few percent at most. Which means anybody can just go out to the bay or go across to the Pacific, take a sample of seawater and make tens of thousands of species discoveries, millions of gene discoveries, et cera. We first, actually, applied these techniques to the human, looking at the bacteria that are associated with us. So the micro biome is your collection of microbes, not all of you have all these cavities, but you know, this is no time for cavity jealousy or anything but we can isolate organisms from these different cavities and for example, look at the person next to you and you know, maybe you see them as having thousands of species in their mouth right now or maybe you can taste your own species. Only about 4 million genes of foreign species in your mouth, in your intestinal tract, if you have those other parts, you have those as well. So we have more bacterial cells associated with us than we have human cells. And it turns out; they affect our physiology in some pretty interesting ways. There’s a company in North Carolina called Metabolomics that is using High-throughput Mass Spectrometry to look at all the chemicals in the blood stream and they’ve worked out that we, as a species, can maybe make around 2,400 chemicals. If you look in the bloodstream of anybody, for example, after they’ve had a meal, easily find about 60% of what’s there is made by our own bodies. About 30% are just chemicals that came out of what we ate. You know, so this notion “you are what you eat†is partly right but 10% are bacterial metabolites so in fact, partly we are what we feed our bacteria and what they give us. But we’re dealing at any one time with hundreds to thousands of foreign chemicals circulating in our bloodstream, not to mention what we add directly by taking pharmaceutics or other things and we have no idea, the impact of these on physiology because, in fact, we didn’t even know they were there before so we’re looking at the complexity of the human genome and all the variations there and what’s in our bloodstream. And if we did this test on everybody in this room, we’ll get a different answer for everybody based on what you ate and also the uniqueness of the bacteria in your own guts, in your mouth. This affects our chemical milieu. This is part of the environment that affects us. We’re also sequencing the air genome. So for example, in a room this size, sitting here for an hour, you would absorb maybe 10,000 different bacteria and maybe 10 times that in viruses in hour. If you go outside, it’s twice that much so you're actually safer in here, right now, depending on who you’re sitting next to, again, and what they’re exhaling and inhaling. So we live in a bacterial milieu. You’re going to the water; you’re surrounded by bacteria so if you’re out swimming and you swallow a mouthful of seawater, you just swallowed millions and millions of bacteria and viruses. The air we breathe, the soil, our own skin, our own cavities so we are as much dependent on bacterial metabolism as human metabolism. We’ve been collecting all these genes, all these information, trying to understand the complexity. And we decided one way to try and understand the complexity was to try and mimic it. This is standard in chemistry to prove that you have the structure that you think you have. You remake that compound. We decided to try and do this with the genetic code starting with some pretty simple organisms The second organism that we sequenced in 1995 was Mycoplasma genitalium. This is its genome, at least, its depiction of it. Very much unlike our genome so it has a little over 500 genes in contrast to our 22,000. We have gaps in our genetic code with no genes that are much larger than this entire genetic code. So bacterial genomes, you can see these bars represent genes, have very little intergenic space. This is still the smallest genome of a self-replicating organism. So, they’ll grow on its own in the lab. There are smaller genomes, but they're dependent on having a symbiote organism for its growth. So, it’s not clear whether they're the extension of the viral world or just truly simple organisms that are dependent on others. And we just ask simple questions, if one cell needs 18,000 genes to live and this one needs 500 is this actually the minimal set. Can we define life in molecular terms based on the genetic code? And so, we set out to try and knock out genes in this genome to see if we could get down to a smaller number. So, every place you see one of these small triangles is where a transposon which is just a piece of DNA very much like a small virus than can randomly insert anywhere in the genetic in the genetic code. Then we select for living cells. So, only genes that are not essential can tolerate transposons going into them. So, this is very much a negative map. You can see some genes can take large numbers of transposons, but keep in mind these experiments only done gene at a time. And we got to a number of around a hundred genes that we could knock out. But we doubted whether if we knocked all 100 out that we could get to a living cell. In fact, if we look at a metabolic map of this simplest of organisms it may look complex, but this is remarkably simple from looking at any one of cells in the human body. And here’s all, if all the different genes that could be knocked out would knock out a lot metabolic pathways probably not resulting in a living cell. So, we decided the only way for it was to actually make chemically in the lab this chromosome. So we could alter its gene composition and that’s how the whole notion of synthetic genomics was born with that relatively simple notion. So we started of with two primary questions, can the chemistry actually permit making these incredibly large biological molecules accurately in the lab and if we make it what can you do with it? DNA is an inner chemical, can we actually boot it up in to a living cell. This is comparing the two areas. So the red line here is reading the genetic code and the blue line is our ability to synthesize DNA. Its actually five orders of magnitude slower right now for writing the genetic code than it is reading it. But this is changing exponentially and basically just over the last few months. And I think that writing the genetic code will soon catch up with our pace of reading it even though it’s still changing. Now, we thought this would be relatively simple. So, there are machines called DNA synthesizers that can make very short stretches of DNA called oligonucleotides. We make pieces that are about 50 letters long and we thought if we just make a number of pieces and they overlap each other that they would just all go together and we could make a larger piece of DNA. It turns out you can but the process with these machines for synthesizing DNA is very inaccurate. It’s a degenerative process, so the longer you make the piece of DNA the more errors there are with it. So, we set out to do an experiment trying to make the Fi x 174 genome. Fi x 174 for those of you who don’t know is a bacteriophage it’s a virus that kills bacteria particularly E. coli. It is one of the first viruses that were extensively studied. In fact, it was the first actual genome of any type that was done by Fred Sanger. We chose this because in fact its genome is very intolerant of changes. If you change the genetic code the virus cannot reproduce. So, we developed some new techniques for actually accurately writing the genetic coed where we can repair the errors in real time as were making it. And for us even though there had been an attempt by another group to make the polio virus which was slightly larger than this. It only had one ten thousandth of the activity because of all these errors and making the genetic code. We went from the genetic code in the computer designing the pieces so that they would go together appropriately use this new process for error correction and we ended up with a piece of DNA exactly the right length. When we sequenced it was exactly what we had designed and were trying to make. The exciting part was we took this piece of DNA and inserted into the bacteria E. coli and what had happened was E. coli recognized this as a piece of software and started making viral particles. And true to form in nature when the viral particles were released from the cell. They turned around and killed the bacteria that had made it. So, this is a process that we see all the time in nature. I was just speaking to oil executives and I said they clearly understood that process. (Laughter) But this was pretty exciting of just taking a piece of DNA and having it activated making viral particles. So we view this as the software actually building its own hardware. This is an important concept as we’re trying to go forward in this field, that even most people that are working in this area have not truly grasped the implications of this, that we don’t have to design life from scratch. We just have to design the software appropriately. In fact we’ve gone to scale this up. Our plan was to make pieces that were viral-sized pieces that we would then out together to make an intact chromosome. We look for a number of ways to do this. We had been working on it for over 4 years and there’s a process in nature called homologous recombination. This is that paper that you might have read about that we just published where we in fact made this entire chromosome. It’s the largest man-made molecule of a defined structure. To print it out, one letter at a time at 10 font with no spaces, it takes 142 pages just to print out the letter code for this structure. Its 582,970 base pairs or letters of genetic code and it’s over 300 million molecular weight. So, the process that we used was not as simple as just making pieces go together but it’s close to that. But it started with the right design. And Design Now is a key part of biology In fact, before we started even the design phase we had to go back and re-sequence the entire genome because the standard in 1995 was roughly one error per 10,000 letters of genetic code. Nobody thought about when reading the genetic code even with that seemingly accurate error that anybody would be using those sequences to reproduce the organisms to actually write the genetic code. And if you’re starting with the genetic code of the computer, you can only make things as accurate as the information you’re starting with. So, we went back and sequenced the genome all over again and found 30 errors, which was the standard for where it should have been. But we know with those 30 errors if we made this chromosome based the initial sequence. It would never been able to be booted up. So we start with this digital information and we design the pieces thinking that we’re now this 580,000 piece into 50 base pair of segments. The 50 base pair of segments have to be designed so they overlap with our neighbor and we started off by making the pieces on the order of 5 to 7 kb. Five to seven thousand base pairs. That each had to be design so when they line up with their neighbor they would overlap. We designed some other unique elements into it. You might be able to see this on the top line that shows the places in the genome where we inserted watermarks. We wanted to absolutely be sure that we would not fool ourselves or others by having a contaminant of even one molecule of DNA from the native organism and this could be the fun part of this process. So, we have a four-letter genetic code. I'm sure most of you know that we have this triplet code were three letters of the genetic code, code for our amino acids. And we have roughly 20 amino acids and there's a single letter code representing each amino acid. So, we can write things in the genetic code using this procedure. And we use this in fact to label the DNA and the team came up with though some of the authors of the genetic code and the institution. We‘ve read that people were very disappointed. We didn’t put any poetry or other more profound statements like one small step or anything like that. But we’ll think about that more carefully for the next time. So, all this went into design started making these pieces, we put four these 5 to 7,000 base pair pieces together to make 24 kb pieces and at each stop we would grow these pieces up in bacteria E. coli to make large amount of the DNA that we then sequence to make sure it was absolutely accurate because not only were we trying to make the end product, which we could have done much faster. We were trying to make absolutely robust methods so we would understand is errors crept in where they came from We combined the 24 kb pieces to make 72 kb pieces and the prior world record for the largest piece of DNA made was around 31,000. So all these pieces at 72,000 greatly exceeded what had been done before. Then we put those together. This looks like a basketball playoff perhaps to make what we called quarter molecules and half molecules. But what we found is exceeded the limits of cloning and bacteria and we switched to another system. We switched to yeast to put these together and to grow them up. In fact, we had always envisioned that we put these pieces together by this process called homologous recombination. Basically, cells and nature do this all the time with repairing their DNA. In fact, this organism Deinococcus radiodurans which is pretty ubiquitous on the planet and perhaps elsewhere can take 3 million rads of radiation. Its chromosomes get blown apart with a couple of hundred little pieces than as long as it’s in water over 12 to 24 hours it remakes its chromosome exactly as it was before. So heres a actual picture of it after 1.75 million rads of radiation. I recommend you not try at home because we as a species can only take a tiny, tiny fraction of this amount of radiation without being killed. We can't do this with our human chromosomes. So this is pretty stunning process it turns out this cell was not unique. We have thousands of species on this planet that can do this. These cells can be totally desiccated. They can be dried out. They can be in a vacuum. They can accumulate this ionizing radiation for a very long period of time. We don’t actually know how long, but the speculation and calculations we’ve done is it would certainly fit in to the Long Now Foundation because it’s tens of thousands to millions of years if the organisms even had slight shielding in a comet or other material. We know this organism can survive in outer space. It reaches an aqueous environment and reassembles its genetic code and it can start replicating again. So, we thought these mechanisms that we’re trying to isolate from this organism would be great for assembling the little pieces of the genome to put it back together again. But in fact, we found it was even simpler than that. It turns out yeast, which is used for making bread and beer and wine all these good things we like can do this on its own with foreign pieces of DNA. So, while we are trying to grow up the pieces in yeast. We found that if we design the pieces correctly and out them in, it would assemble those automatically until the larger pieces into the intact chromosome. So that’s how we ended up with the entire 580,000 base pair piece that we sequenced and it’s down to 0 errors. This is actually a picture of it. You don’t need an electron microscope because this molecule is so large. This is just looking at over a 6 second period as you can actually see it’s a circular piece of DNA. It’s pretty exciting for us to be able to actually visualize it. Well, how do you boot up a chromosome? You saw with the virus all we had to do was insert it in a bacteria and the bacteria can start reading that software and producing things. We think it’s a little bit more complicated with whole bacterial chromosomes. Initially, we thought that we would have to try and remove the chromosome from a bacterial cell and add in this new to replace it. It turns out that’s very hard to do. So what you’ve heard about with mammalian cloning works very easily in eukaryotes where there’s a defined nucleus that’s easy to just to cut out of the cell under a microscope. Lift it out and out in a nucleus from a different organism or a different cell from the same organism. With bacteria and archaea, there is no nucleus. The DNA is part of the cytoplasm of the cell so we have to use a little be more ingenuity to do this and last year we published what we think is the key technique of transplantation that’s pretty stunning in its own right because we, actually, by putting in a new chromosome into a cell that had a different chromosome in it to start with. We completely converted one species into another so it was an absolute 100% transformation. We used two closely related mycoplasma species roughly the same distance apart as we are to mice. We isolated the chromosome from this M. mycoides cell and we wanted to make sure we can get down to just DNA. Chromosomes have a lot of protein associated with them so we treated these pretty harshly with digestive enzymes to digest away all the proteins. And then we added back a few additional genes to this chromosome. One set of genes, the lacZ set, for example, will turn the cells blue so it makes it easy to identify them. And we also added in a set of selectable markers so we could select for cells just for this transplanted chromosome and we put this chromosome into a species called a capricolum. Now, here’s a wonderful graphic. I'm sure you all appreciate the sophistication of this. We transplanted the chromosome into the cell. In fact, we thought we would end up with this situation, a cell with two chromosomes in it. We see this all the time in nature, so all these people that make arguments against evolution because we know that you can't just get a point mutation in one piece of DNA unless it’s for the wavelength of light your photoreceptor sees but that’s going to change into more complexity. What we see in the real world is we see chromosomes moving around where you can add a thousand new traits to a cell in a second. Cholera, for example, most people thought there was no point in decoding the cholera genome because it was very closely to E. coli but when we read the cholera genome it turns out that it didn’t have just a single chromosome. It had two chromosomes and it looked like this. They were very different from each other. Obviously, it had taken up a chromosome from another species and added it to its repertoire. But we didn’t want this situation. It turns out the capricolum genome is very unusual in that it doesn't contain any genes for restriction enzymes. Restriction endonucleases are the molecular scissors that cut up DNA and it is how cells protect themselves from this foreign DNA coming in. In fact, the chromosome that we put in did have a restriction enzyme. As soon as it was in the cell, it got expressed and it recognized the original chromosome as foreign and cut it up into small pieces and it got digested. So were left with cells just with the transplanted chromosomes In over a very short period of time, we ended up with these blue cells that all the characteristics of these cells were that what was dictated by the transplanted chromosome. Every protein in the cell changed from that with the original species into that coded by the M. mycoides chromosome, the membrane changed, everything changed. We can isolate the DNA and it was only what we had transplanted in. So this is true identity theft at the ultimate level and fortunately, either most of us have mechanisms to protect us against this, and most cells do with these restriction enzymes but they give us very powerful tools to try and do this in the future so we know we can take a chromosome. We can transplant it. We can completely convert one species into another. So we’re in the process of doing this right now with the synthetically made chromosome. We hope this will happen very soon. These experiments go very slowly because the cells grow slowly. So it takes about 6 weeks once you do a transplant to see whether you have viable cells but we should be seeing that, hopefully, in the very near future. That will complete the trilogy that we had been putting together on this, although, it’s clear now just from the transplant experiment that we know this will work and to me, it’s more of a technicality if it works with synthetically-made DNA versus DNA out of a cell because it is the same – it sequences the same but I think it’s still important for the proof. But it transforms us into a new era of now (indiscernible)[00:51:47] alter cells by simply rewriting the genetic code. So the next steps and why. I'm going to walk you through quickly. Some of the reasons why were doing this and where were trying to go. But a lot of people when they think about this and write about it, they think we’re redoing Genesis and we’re not. This is much more like a new version of the Cambrian explosion. We’re dependent as the title of this lecture said on the 3 1/2 billion years of evolution. Also in because we’re as humans because of that and we’re using organisms that have evolved and substantial period of time. What’s different is that we don’t have to redo all of that because if we just write new molecular software, we can start at that point and go in an infinite number of directions. Why do this? One of the reasons is trying to deal what the future of our species on this planet over the next 40 years we’re going to go from 6 1/2 to 9 billion people. That’s a huge change in a short period of time. I try to put this data in context for myself and hope that it leads others to be able to understand it. So I was born in 1946. There's now three people on this planet for everybody that existed in 1946. So if you think of having 1/3 of the number of people in this room in 40 years of before people for everybody that was on the planet in 1946. We’re having trouble providing food, clean water, energy, housing for the 6 1/2 billion. We’re going to have even more trouble trying to provide it for 9 billion people. We’re changing our environment quite dramatically by burning billions of years of biology in the form of oil and coal. This is now from 2003 so, you know, over 5 billion tons of coal. Billions of barrels of oil that we just take out and burn and the CO2 goes into the atmosphere. The slide is out of date. These should be changed again the third time in a year. This number is now 4.2 billion tons of CO2 that were adding and stays in the atmosphere each year. That number is accelerating. The ocean sink which is the largest sink as you can see that they can take hundred of billions of tons of CO2. Many people think it’s saturated and deforestation is also contributing as we try and strip more land to make more food and fuel everybody who have seen this graph and the steady increase of CO2. So what can be done? There is a dramatic revolution that’s taking place in the industrial world equivalent maybe to the first industrial revolution. Companies like DuPont which was basically built on using oil as its raw material, had now had to switch away from using oil. They are switching to sugar which maybe won’t be anymore sustainable and well talk about that in a second. But they spent 10 years but over 100 million dollars offering E. coli to do a simple reaction. To make 6carbon sugar and make 3 carbon propane dial molecule. And if you can see those what looks like a large beer kegs, those are actually 4 6,000 liter fermenters that they grow up literally 100 tons of this bacteria at a time in a batch and make propane dial from sugar with this engineered bacteria. They can do this cheaper and faster than they can with chemical conversion with their chemists. This has certain downfalls after each batch. They have to bury those tons of E. coli in a land fill because they are not allowed to burn it. But even all these limitations and the cost of sugar they claim that their new polymer that uses this chemical will be the first multi-billion dollar biotech product that’s not a pharmaceutical. Some metabolic engineering is being used. My new company Synthetic genomics has a deal with VP to try and use biology deep in the Earth to stop mining coal by biologically that coal into methane. We were quite surprised, a mile down in the Earth when we took the first sample. There was more biological diversity there than we found in the ocean. Just a simple microscopic field was teeming with microorganisms. This is a piece of coal that we have in laboratory. All those little tiny things or the bacteria that live off a coal was substrate and we have other bacteria that convert that substrate right into methane. It’s a little bit dark but maybe you can see the methane. Bubbles coming off a coal in the bottom. This doesn’t stop taking carbon under the ground but it’s about a 10 fold improvement over mining cold and burning it. And the coal reserves are extensive enough that this could maybe get ourselves a stopgap until there is a new economy. But we’ve been going in the wrong direction already and part because of lobbies in the government corn to ethanol. Just does not going to get us there, it’s a negative carbon balance it’s been heavily subsidized by all of us. And the numbers are pretty stunning it looks like a huge amount of ethanol were being produced and we actually look at the numbers on the order of 6 billion gallons. There's now I think 160 plants around the country, 6 billion sounds like a lot until you look at what we use. So this is our transportation fuel, 140 billion gallons of gasoline, 45 billion gallons of diesel. So 5 to 6 billion gallons of ethanol basically does nothing for this equation. Especially when it produces more CO2 than it captures. It’s competing with farmland. Food prices are going up because of this. Corn prices have doubled on the last year. So this is just the wrong experiment taking us very much in the wrong direction. Now, were not short of energy on this planet. A hundred twenty thousand terawatts of energy arrives here each day from the sun. If we could capture that at 1% efficiency, we’d only need about 5 billion hectares. I’ve argued this is a great use of Nevada. Nobody has ever argued with it. They haven’t come up with better alternative views. We could just from sunlight produce energy. In fact, deserts and seawater don’t compete with food. We’re in fact working on, I’ll show you in a minute what were calling a fourth generation fuel starting with CO2 as the feed stock. So if we can start with CO2 either from sunlight in the methanogenesis pathways. We skip all this use of farmland for producing fuels. We have two fuels that are pretty close going into test production we call them second and third generation. They both have the downfall of using sugar as a starting material but they're far superior to ethanol. They don’t mix with water. They have very low freezing points they can just be fluffed and blended they can go through pipelines they have a much higher energy density to get people going in the different direction, but the one that we think we’ll have in about 18 months, the fourth generation fuels, starts with CO2 as a feed stalk. Now, if we can really use CO2 as a feed stalk it does several things when you consider the efforts for CO2 sequestration. We can just take piped CO2 into reactors, biological reactors and make fuel such as octane directly from carbon dioxide, either using sunlight as the energy source or molecular hydrogen. You won’t see any energy plans incorporating this types of ideas. We’d go forward by extrapolating from the past from linear ideas or linear thinking. There is no planning for disruptive technologies, but in fact if were going to change in anything in anybody’s lifetime here, we need disruptive technologies at the Davos meeting this year the conclusion was 40 years from now biology and all alternate energy sources will maybe only make a few percent impact on using oil and coal, because there's so much vested interest. There's a new cold powered fuel power plant coming online everyday. These caused billions of dollars to build. So, it’s only if there’s an economically viable alternative that doesn’t compete with food that has a chance to change anything in the near future and I don’t think we want to keep adding CO2 to the atmosphere. So we have 10 million genes in the databases right now. We’re going to double that number again this year from the Sorcerer expedition to 20 million. These are our design components. The entire electronic industry only had a few handfuls of designed components and look at the diversity that we have there. We have tens of millions of starting components that we can help. And now we built a robot that instead of making one chromosome per 4 years. We think we can get up to making thousands to millions to millions of them a day and in random combinations. So were calling this new field combinatorial genomics. So just think of it, if we can basically start with design take these tens of millions of design components. Just make the DNA transform that automatically in to yeast or some other vector and screen for chemical production for octane production antibody production. Whatever you screen for will be the selection that happens. It will change the pace of biology in unimaginable ways. Just like I don’t think anybody could imagine in a 1950s this laptop computer went close to terabyte of storage on it doing more than giant rooms could. This is a new area that’s going to be driven by people’s imagination. This is sort of software that were working on. To make the new software thinking of this as a true design phase. Trying to come up with new fuels, new chemicals. And so it’s an exciting time as we try to scale this up with remarkably small teams. This is a different aspect of industrial revolution. And that it doesn’t take armies of thousands. In fact that’s what we’ve done from the beginning with reading the genetic code. The yeast genome took 10 years and a thousand scientists. We reduced that project first to 4 months and now to ours and were doing the same thing now going in the other direction. It’s clear we’ll be able to have a huge dent in the first bullet here in increasing the understanding of life because we need to use empirical methods with the huge data sets that we have. There's not enough scientists to ask reasonable questions to work out the biology of what’s been discovered to date. But I’m more optimistic I'm hoping within 15-20 years we can really have gone a long way for replacing the petrochemical industry. Hopefully, biology would be major source of energy. And bioremediation if we can capture back CO2 and recycle that into fuel. We possible even start to undo some of the damage that we’ve been doing for the last 100 years. And were also looking at this in terms of every type of chemical and looking at making for example combinatorial anti-bodies so the we can have instant vaccines for every variant of the flu virus for example instead of being a year late like we are again this year. This field is unusual and this is my last slide by the way. It started with us asking the ethical questions before we did the first experiment not after we thought “Gee whiz, we have a lot of potential here.†The result of this team at the University of Pennsylvania published their review in 1999. They thought we were proceeding with the right questions and the right technology. There was a caveat about being concerned about biological warfare. The Sloan Foundation just funded my institute and MIT for the last couple of years to do a complete review of the risks and the benefits. I think I published at the end of last year and there are ongoing discussions a new body in the federal government got formed when we published Fi X 174 genome. Got reviewed at the White House, they had to decide whether to try and make our data and processes secret. I think its one of the few good things of the Bush administration has done in sciences. They went and opted for open publication. But they formed a new branch, executive committee that has people from every branch of the government reviewing this type work we’ve had public discussions ongoing basically since 1996 on this. We’re trying to take it one step at a time but it’s about to expand in an exponential fashion and I think its going to be an exciting next 10,000 thousand years. Thank you very much.