Begun in 2008, the "1000 Genomes Project" aims to sequence 1000 genomes and gain a deeper understanding of what genetic variations may put people at risk for disease.

When the Human Genome Project got underway in 1990 it was expected to take 15 years to sequence the over 3 billion chemical base pairs that spell out our genetic code. In true Moore’s Law tradition the emergence of faster and more efficient sequencing technologies along the way led to the Project’s early completion in 2003. Today, 22 years after scientists first committed to the audacious goal of sequencing the genome, the next generation of sequencers are setting their sites much higher.

About a thousand times higher.

The 1000 Genomes Project, as its name suggests, is a joint public-private effort to sequence 1000 genomes. Begun in 2008, the Project’s main goal is to create an “extensive catalog of human genetic variation that will support future medical research studies.” The 1000 Genomes Consortium is headed by the NIH’s National Human Genome Research Institute which in turn is collaborating with research groups in the US, UK, China and Germany.

That might not sound like much. Thanks in large part to companies like Silicon Valley start up Complete Genomics perhaps as many as 30,000 complete genomes around the world have already been sequenced. But what is unique about the 1000 Genomes Project is that their genomes will be made available to the public for free, and stored in a place where the world can access the data easily and interact with it.

The original effort to sequence the human genome, while a triumph, is limited in its usefulness insofar as linking genetic sequence to disease. Because it involved DNA from just a small number of individuals (the fifth personal genome, that of Korean researcher Seong-Jin Kim, was completed only in 2008) it is impossible to use the data to make correlations between genetic variations and diseases. The 1000 Genomes Consortium hopes that their sample size will be large enough to catalog all genetic variants that occur in at least 1 percent of the population.

Among the 3 billion base pairs contained in the human genome scientists have already identified more than 1.4 million single nucleotide polymorphisms, or SNPs (pronounced “snips”). SNPs are single base variations that differ between people. By characterizing which people have which SNPs, scientists hope to identify the SNPs that predispose people for diseases such as cancer or heart disease. Smartly, the Consortium is not limiting themselves to any particular population, which might bias the genetic variability to disproportionately represent that population. The equivalent of 1,000 genomes will actually be gotten from the incomplete sequences of 2,661 people from 26 different “populations” around the world.

Understanding which single nucleotide polymorphisms increase risk for disease will not only help treat the disease, but may also contribute to a cure.

Just as advances in sequencing technologies throughout the ‘90s galvanized the Human Genome Project, advances in the last decade have put 1000 genomes within reach. So-called “next-gen” sequencing platforms reduced the cost of DNA sequencing by over two orders of magnitude in just a three year span. The lowered cost meant that individual labs could get in on the sequencing act and contribute to the kind of large-scale sequencing that had previously been the domain of major genome centers. And not only was more data being generated, but techniques to verify the quality of the sequences significantly improved.

Sequencing technology will undoubtedly continue to improve until the next “next-gen” sequencing platforms will allow us to sequence even faster and more cheaply. But the current swell in DNA data has put pressure on another technology to keep pace.

The amount of data generated from DNA sequencing is prodigious. Right now the Project has already amassed over 200 terabytes of data. That’s equivalent to 30,000 standard DVDs or 16 million file cabinets topped with text. According to the NIH, it is the largest set of data on human genetic variation. Not to be overburdened by a few hundred terabytes, Amazon announced last week that the 1000 Genomes Project data is now stored on their Amazon Web Services cloud and is publicly available. It currently contains sequence data from about 1,700 people. Sequencing the remaining 900 or so samples is expected to be completed by the end of the year.

You can find information on how to access the data here.

As if to answer the call for improved data handling tools, the Obama Administration last week launched its “Big Data Research and Development Initiative” that basically spreads $200 million across six federal science agencies to fund R & D of technologies that “access, store, visualize, and analyze” enormous sets of data. The 1,000 Genomes Project is part of the White House Initiative.

The National Human Genome Research Institute, rightly so, calls the Human Genome Project “one of the great feats of human exploration in history – an inward voyage of discovery rather than an outward exploration of the planet or the cosmos.” For the first time we were able to map our entire genome from end to end. Our estimate of total genes was whittled down to between 20,000 and 25,000, we have a better understanding of our relatedness to other species, and not to mention, we’ve discovered gene mutations associated with breast cancer, muscle disease, deafness, and other illnesses. Who knows what the 1,000 Genome Project is yet reveal about our DNA and ourselves. We’ve painted a digital portrait of our DNA, we now begin to add the finer strokes.

[image credits: National Geographic and DNA Sequencing Service]
image 1: DNA array
image 2: drugs

Old friends and new allies poured $22 million into 23andMe. New research may be the key to keeping the company secure.

23andMe, one of the pioneering companies of personal genetics, is ramping up for another round of growth. According to a recent press release, the Mountain View based DNA analysis business raised more than $22 million in Series C funding. While it’s no surprise that early investors Google Ventures and New Enterprise Associates continued to back 23andMe (Google founder Sergey Brin is married to 23andMe founder Anne Wojicicki), the arrival of newcomer Johnson & Johnson Development Corporation shows that the startup is still attractive to outsiders. Wojicicki commented that the $22 million would go a long way towards expanding their research efforts. While 23andMe has had its issues in the past, this new round of funding could be a sign that the company is ready to bounce back stronger than ever.

Companies like 23andMe test the DNA of customers and analyze it for important single nucleotide polymorphisms (SNPs) – a process that has drawn fire in the past year for its over simplifications. 10 years after the first human genome was sequenced, genetic analysis has yet to provide clinically proven benefits to the majority of customers who use the services of companies such as 23andMe. The FDA, US Congress, and other government regulatory bodies are actively investigating how best to control mail order DNA testing. Experts, like Craig Venter, have dismissed the current state of personal genetics almost entirely. How much can we really predict about your health from just a few random variations in your DNA? Analyzing SNPs may be too simple of an approach to understanding the importance of your genes.

Yet clearly 23andMe’s investors are making a $22 million gamble that SNPs will prove to be a valuable commodity in the years ahead. It could pay off. At the recent annual meeting for the American Society of Human Genetics, 23andMe presented some of its ongoing research in linking SNPs to important medical conditions. Last year, they set out to gather data on 10,000 Parkinson’s patients, and have managed to collect samples from 3500 so far (with an additional 20,000 controls). 23andMe has also replicated 100 known associations for genetic conditions and found genetic links for the Mumps. All of this is part of their more general project of democratizing genetic research through customer data collection and feedback.

I don’t know how personal genetics will evolve in the next few years. If I had to take a guess it would be that whole genome sequencing, which is steadily dropping its price towards the $1000 level, will become the new gold standard. Where this will leave SNP analysis and personal DNA testing companies no one can say. It’s an uncertain time, but I think that 23andMe is making the right call. The only way they’re going to survive is by increasing our understanding of genetics. Research is the key to validating the personal genetics industry. We’ll see if $22 million is enough to make that happen.

[image credit:23andMe (modified)]
[source: 23andMe blog and press release]

Venter, master of DNA, speaks bluntly to Der Spiegel about genetics.

“We have learned nothing from the genome.” That’s the grim message that J. Craig Venter recently gave Der Spiegel in an amazing interview. Venter, decoder of the human genome and creator of the world’s first fully synthetic bacteria, doesn’t pull any punches when describing the medical benefits we’ve derived from sequencing the human genome. They are “close to zero to put it precisely.” The Der Spiegel interview catches Venter in a blunt mood and we’re given a rare insight into how one of the foremost scientists in the field (probably the foremost scientist) see our progress thus far and our hopes for the future. To paraphrase: we haven’t really accomplished anything yet, people don’t want to believe that at all, and we’re finally taking the first steps to really understanding things now. Some of Venter’s juicier statements have me rethinking the current state of genomics. Check out the quotes below.

As we recently discussed, the 10 year anniversary of the Human Genome Project (and Venter’s competing and more successful Celera Genomics) has raised serious questions about what we have really learned from our foray into genomics. We’ve had success with in vitro screening for certain genetic illnesses, and we’ve used genetics to craft a few new medications, but the public at large has not seen a lot of benefit. Why?

Because we have, in truth, learned nothing from the genome other than probabilities. How does a 1 or 3 percent increased risk for something translate into the clinic? It is useless information.

We’ve seen personal DNA testing become a burgeoning product, with companies like Pathway Genomics, 23andMe, and Navigenics offering to scan your genome for important genes (SNPs) and tell you what their presence may mean. Such tests are all about probabilities, and many have raised the same concerns as Venter – that we what we learn from such studies is practically useless. I for one, enjoyed my DNA test, but mostly for the educational merit. I’ve haven’t changed a single habit in response to the data I was given. Why?

we need a lot more information: Information about your body’s chemistry, your physiology, your complete medical history, your brain and your entire life. We would need to do that a million times on different people and correlate that data with their genetic information.

Essentially we have more DNA than understanding. Until we can correlate massive amounts of genetic data with real-world effects we really don’t know what to tell people when we give them results to personal DNA tests. Even those that have had their whole genome sequenced (not just SNPs) don’t really have much insight into their lives. As Venter says about his own experience with sequencing: “We couldn’t even be certain from my genome what my eye color was.”

But efforts are already underway to correlate medical histories with DNA. We’ve already discussed biobanks of hundreds of thousands of patients that are being created over the next decade. With cheaper whole genome sequencing it will eventually be possible for us to examine this rich pool of data and perhaps discover meaningful and useful insight into how genes affect our bodies and health. What happens then?

It’s not, ‘Oh, we know your genome, we’re going to make this drug for you.’ That will never happen. It is more important that you use the information in the genome about your personal risks and reduce them through intelligent behavior.

I heartedly agree on the last part, but object to the first. We don’t know if it will be practical to tailor drugs to individuals (probably cost prohibitive) but I think it is likely that we’ll be able to customize treatments. Doctors already do that for every patient without a lot of knowledge about genetics. In the future we may not create entirely new medications for each patient but we’re very likely to use fast chip processing to give doctors an idea which drugs will react poorly with the patient due to genetic predisposition. Given enough possible combination of medications the difference between custom making drugs and custom designing a cocktail of them will seem slight. In my opinion.

If you think that Venter’s comments in Der Spiegel are unbearably gloomy, you haven’t read enough of the interview. Yes, the first third is mainly Venter trash talking about Francis Collins, and the second third chastises everyone for thinking that genetics was some magical cure-all or some dreaded infringement on God’s turf. The last bit, however, reveals where Venter’s hopes seem to lie: in creating new life from scratch.

We don’t even know how the simplest bacterial cell works. We want to learn what the minimum cellular components are, so we’re going to be taking out all the non-essential genes.

He’s already assembled bacteria from the building blocks of DNA, and now he has his sights on using that technology to really advance our understanding of genetics. He and his team plan on building a ‘minimal cell’ – the simplest form of bacterial life you can make and still have survive. The hope is that this will lead to a greater understanding of what it takes to be an organism – to understand the basic components and operating system of cellular life.

From that understanding could come the ability to truly design organisms from the ground up. We could design bacteria that produce complex carbon compounds and reduce our need for oil. Exxon Mobile has invested $600 million with Venter in the hopes of creating these new life forms which could radically alter the availability of resources all over the world. Think of what the organisms we design may be able to produce.

Not only gasoline. Plastic, asphalt, heating oil: Everything that we make from oil will at some point be made by bacteria or other cells. Whether that is in five, 10 or 20 years is unclear. Why don’t we have fuel now other than alcohol from microbes? It’s because nothing evolved that can produce great amounts of biofuel out of CO2. That’s why we have to make it.

Venter’s vision of the future seems to be as hopeful as his vision of the past is disdainful. There is so much power in genetic engineering and synthetic biology that it’s hard to fully grasp both its limitations and its potential. If there is one message to take away from Venter’s comments I think it is this: we need more understanding. Thankfully there are many, including Venter and his colleagues, who are pursuing that understanding with a fiery passion.

All quotes are J. Craig Venter,2010 as taken from Der Spiegel. The full Der Spiegel interview with J. Craig Venter conducted by Rafaela von Bredow and Johann Grolle can be found here. It is awesome. Read it.

[image credit]
[source: Der Spiegel]

Two genes help Tibetans live on the Roof of the World.

Researchers from the University of Utah and Qinghai University in China have discovered two genes which assist ethnic Tibetans survive in the high altitude of the Himalayas. Their study, recently published in the journal Science, compared key genes in 31 unrelated Tibetans to those of 45 lowland Chinese and 45 Japanese people. They found that the EGLN1 and PPARA genes were responsible for the characteristic lower levels of hemoglobin in Tibetans – revealing half the genetic mystery in to what allows them to survive in the thin air. Further research in this field may not only reveal the genes responsible for high-altitude adaptation, it could be the first step in providing the same genetic boon to people all over the world.

Tibetans are not the only ethnic group in the world which has adapted to high altitude conditions, but they do seem to be the only ones that have done so with a low red blood cell count. Most other groups, including mountain-dwelling groups in the Andes, compensate for lower oxygen levels by increasing hemoglobin levels. This does the job very well, but extremely high hemoglobin levels do have associated health risks (i.e. issues with circulation). Tibetans actually have lower than normal hemoglobin levels, indicating that they are using what little oxygen is in their blood very efficiently. By determining which genes cause this lower hemoglobin level, scientists have found the first (perhaps less exciting) half of the Tibetan genetic puzzle.

The Chinese and American researchers isolated the EGLN1 and PPARA genes by carefully narrowing down possibilities. From databases of genetic samples in China, researchers were able to determine 247 genes that were likely to code for oxygen processing. Then they examined the DNA of 45 Chinese, 45 Japanese and 31 unrelated Tibetan individuals and found those variations which were present in Tibetans but not the other two groups. From there, they narrowed the list of possible genes down to those that had the greatest frequency in Tibetans with the lowest hemoglobin counts: EGLN1 and PPARA.

While this study was very thorough, it’s important to note that the research did not use whole genome sequencing. Rather, a SNP array looked at 900,000+ SNPs (single nucleotide polymorphisms) to find the important variants. Data for the Chinese and Japanese individuals came from the HapMap project and were collected before the Tibetans. None of this limits the results of the study, but given the complexity of the gene interactions there may be more to find out using a whole genome sequencing technique.

I should also point out (again) that lower hemoglobin levels are just the first half of the genetic puzzle. And, to be blunt, probably the less useful half. We don’t have a huge need for lowering the oxygen levels in people (at least for now). What we really need to know is how Tibetans are able to live with such low O2 levels and survive. What’s the genetic secret behind the associated oxygen processing efficiency that comes with their lower hemoglobin levels? That’s the second half of the mystery and it will hopefully be solved by the continuing work of these researchers.

We progressive technological types tend to want the best of everything, and genetics is no exception. Finding out which genes code for better oxygen processing in Tibetans is a great scientific question, but I think it’s exciting because finding the answer could lead the way to giving the same benefits to non-Tibetans. Other scientists are trying to find the genes that code for life-long athleticism, and the genes that promote rapid muscle growth. Taken together, this research could lead to optimizing the human body for performance by either changing genes, or (much more likely) finding ways to promote the protein pathways controlled by those genes. In the future we may all be able to pop a pill and run around the Himalayas as if they were Miami Beach. Someone tell the sherpas to start looking for new jobs.

[image credit : iStockPhoto via Science News]
[source: Science News, Simonson et al, Science 2010]

Brin appears to have sunk $10 million into this series B round, while Google has put in $2.6 million.  As if the financial investments weren’t controversial enough, apparently Google and 23andme have entered into some sort of leasing agreement, though the details of this agreement are not available.

For those with their heads in the sand, there are two major types of personal genome sequencing out there.  In the first type, your entire genome is sequenced – every single one of your 3 Billion base pairs.  This procedure is expensive and time consuming, and although companies like Complete Genomics are poised to bring this ability to the masses for about $1,000 per individual in the next year or two, the price is currently much higher.  For the masses who cannot wait for the full genome sequencing from the likes of Complete Genomics, an alternative is to have more than 1 million of the most important or interesting chunks of your DNA, called SNPs, analyzed for much less than $1,000 today.  There are two major players in this space, 23andme and Decodeme.  As we reported earlier, Decodeme is facing imminent bankruptcy, and this latest round of funding shows that even for industry front runner 23andme the market is a tough place to be.

Although the future for personal DNA sequencing is eventually going in the direction of full sequencing of every single base pair, 23andme offers a valuable service in the near term that is charting new ground and helping to pave the way for the ongoing genetics revolution.  Given that the future seems to be in whole genome sequencing, rather than with SNPs, the long term future of 23andme seems perilous…I would not want to be one of their investors for the long term.

Ethical investments and economic viability of the company aside, you gotta hand it to 23andme for being an innovative and leading company in the field of genetics.  They have really stirred things up: making genetics cool, bringing real genetic tools to the masses, and proposing bold initiatives for conquering disease.  Most notably perhaps, our earlier story on the 23andme initiative to gather DNA samples from at least 10,000 people as part of a massive effort to identify genes that may be at the root of Parkinson’s disease is laudable.

Now with more money in the bank, 23andme should be able to plug along for another year or more and further the genomics revolution.  Lets wish them well, for we need all the help we can get to achieve the promise of genetics.

(Disclosure: In my previous career I worked at Google)