YouTube Facebook LinkedIn Google+ Twitter Xingrss  



Horizons 
 

Nov. 12, 2002 | IT is fitting that John Craig Venter's boat is named Sorcerer II. The 56-year-old California surfer-turned-scientist has long relished the role of maverick wizard sent to bedevil the genomics establishment. Venter cast a contentious spell over the international effort to sequence the human genome that still hasn't lifted. His sequencing savvy and entrepreneurial élan never sat well in polite research circles. More than a few competitors smiled smugly when Venter was unceremoniously fired as CEO from Celera Genomics Group last January. Nevermind that Celera did, at least in part, what Venter claimed it would — produce a thoroughly usable sequence of the human genome, years ahead of schedule.

Now he's back, unbowed and full of new ambition. He has created three nonprofits — The Center for the Advancement of Genomics (TCAG), the Institute for Biological Energy Alternatives (IBEA), and the J. Craig Venter Science Foundation — to complement his original effort, The Institute for Genomic Research (TIGR). There's plenty to do, he says, like the kind of science he was unable to do while running a company.

But first, there's some unfinished business: What really happened at Celera? Why the lingering animosity with the public Human Genome Project? What of the future of genome sequencing, drug discovery, and information technology? In an exclusive interview conducted during TIGR's recent Gene Sequencing and Analysis Conference in Boston, Venter sat down for two hours with Bio·IT World editor-in-chief Kevin Davies, and executive editors John Russell and John Dodge, to discuss the past, present, and future of genomics — and much more.

In Part I of a two-part series, Venter talks frankly about the race for the human genome, his firing at Celera, his detractors, and the financial problems facing the biotechnology industry. Part II, published next month, includes his plans for the future, including the possibility of teaming up with Microsoft.


Q: A lot of people were surprised after you left Celera in January and didn't jump straight into another company. What happened?
A: In my entire career, I've only spent three years in a company. The whole reason I invented the HGS-TIGR [Human Genome Sciences-The Institute for Genomic Research] structure was because I didn't want to be in a biotech company. Most companies set up a research group and the founding scientist heads up the research teams. But the notion I've seen over and over again was the founding scientist always gets killed off. I call it the "praying mantis syndrome" of biotechnology. As soon as you fertilize the company, you get your head bopped off!

So I thought I was being very clever. I formed an independent not-for-profit institute so they [financial backers of HGS] couldn't kill it. They still tried. And my one venture into Celera — people tried to put all kinds of nice terms on it — but I got fired! I was fired from Celera because I said I was leaving in another six months. So it became, 'You can't leave; we're firing you!'


Q: Were you relieved when you got fired? Angry? Hurt?
A: The way it was handled was probably as tacky as anything could ever be done. I had decided to leave, and one of my worst characteristics is I always telegraph my punches. I never sucker-punch people.


Q: You've said that running a company was merely "a means to an end." Is that because of the sudden way it ended? Did you start Celera thinking, 'I'm just going into this for a short time and then get out?'
A: Yes — and Tony White, the CEO of then PerkinElmer, knew that because I had set up two goals at the beginning. One was the sequencing and the genome. The other was to build the endowment of my foundation so I could go back and do science when it was done.

I felt you don't enter into these Faustian bargains without holding up your end of it. They were putting up the money, so I had to play by their rules. I'm the one that invented the [bio-content] information business, and the Applera people never liked it. They never understood it. They never supported it. Even when it made money and became profitable, they went into denial over it.

Tony White stated that his goal was to sell more instruments. And it worked. It's not clear in retrospect whether Celera was supposed to be successful or not.


Q: Why was Celera so successful initially, and why was it forced to change its business model?
A: It was the peak of irrational exuberance. It was stunning that the Celera market cap went up to $14 billion. Instead of business people being rational and saying 'This is crazy,' [they said] 'You have to change what you do to justify keeping that market cap.' The information business was never intended to do that the first couple of years.

I think [Celera] is one of the few successes, however briefly, in biotech, and certainly it has to be the best story in bio-information technology. Even though some of the data were available for free from the Web from stuff we put out, [and] from the public [Human Genome Project], scientists around the world paid for quality informatics, quality programming, quality compute capabilities, and just quality data.


Q: Should these data be available commercially? Is it one rule or the other, or should it be in the public domain?
A: We put the raw sequence data in the public domain. It's on the [Celera] Web site. Any researcher can get a DVD for free just by requesting it. What I put out as a model was that the raw data could be available anywhere for free. The public sequencing gurus totally misstated that, misrepresented it totally.

I think Celera's success commercially actually proved that the value was in the interpretation, the tools, the computing capacity, etc.


Q: Just to clarify, when the February 2001 Science human genome paper came out, the information was available to academics — but weren't there restrictions on the amount you could republish and how much you could access?
A: There's no limitation on how much anybody could get access to, because they could get the whole thing on a DVD. The only limitations were that somebody couldn't take the data and republish it. You can't copyright that data. You'd probably be upset if I took your magazine and republished it on my Web site. All we said is that you can't take our data and republish it on your Web site, because the business investment behind it is what paid for that data. Now, if the taxpayers pay for the data, the rules are totally different.

There's not a single scientist who's been responsible for putting more data in the public domain than I have. We've put hundreds of thousands of genes in the public domain, more genomes than any other group, and it totally depends on who funds it. I call it the golden rule: He who has the gold makes the rules. The bizarre story is [it's] the people that get these tens of millions of dollars in grants and public funding that want to keep the data back.


Q: Since the human sequence came out, wasn't the rice genome sequence published by Syngenta in Science this year under a similar model?
A: Yes. In fact, it was the identical model. They used the Celera agreements with Science and copied them.


Q: So this could continue for other potentially commercially important genomes. Isn't that going to hold science back?
A: In fact, it moves it ahead. We have the human genome and the rice genome now because of those private investments. So whether I'm in academia, which I am now, or when I was in the commercial arena as a citizen, my view is anything that drives [science] forward is good. We're all better off because we have access to the rice genome data today instead of waiting for some 15-year public program.

Even my worst critics ... they all confessed that we moved things forward. I would prefer to have done all this in one of my not-for-profit institutes. I haven't quite worked out how to get somebody to give me $350 million just to do the experiments I want to do. So I think the model was the perfect compromise. In fact, it's a compromise that I forced on the industry.


Q: There seems to be a stampede by genomics companies, including Celera, to reinvent themselves as drug discovery companies. Do you think they're going to be successful?

"I got fired! I was fired from Celera because I said I was leaving in another six months. So it became, 'You can't leave; we're firing you!'" 
A: I don't know. It's hard to predict, but most of them will fail. The only thing Celera really has going for it is the cash that I raised and left there. And that was raised on honest pretexts. Nobody had any idea it would end up being the largest fund ever — a billion dollars for a biotech company in one week for 5 percent of the company! But when the powers that be were trying to appease people that wanted a $14-billion market cap sustained, [that] means you have to have billions of dollars in revenue. So with this total shift and false expectation, all of a sudden there's no way information technology, in its current form ... could ever live up to that.

So it was a self-created failure. Instead of being smart and understanding that the markets were totally out of whack, [people concluded] what we were doing had to be out of whack because we weren't making billions of dollars a year. Amgen barely makes billions of dollars a year with real pharmaceutical products that they developed over a 15-year period.

That's where the dishonesty of the system comes, first in genomics, and now in drugs. They'll go anywhere that Wall Street analysts tell them to go. Could I guarantee that we could make drugs? No, because I think it's extremely hard to do that. Not only do you need scientific breakthroughs, you also need a hell of a lot of luck to do that by itself. It went in that direction totally because the heads of the organization felt that information technology could not drive multibillion-dollar revenues instantaneously.


Q: Is one lesson that scientists shouldn't be CEOs?
A: No. I think being a CEO is actually [something] more scientists can do, better than most businessmen. Business is not difficult. It's pretty intuitive. It's pretty logical, if you're doing it honestly. You either have something that really you can sell or you don't. If you don't, then you enter into this realm of the BS and dishonesty that part of the commercial world does. You have to have a vision. You have to know where you want to go with it. ... If you don't have that and all you have is a pool of cash and a whip, you're not going to do anything.

Celera had something bona fide to sell; we just had people that didn't understand that business. When you go from groups that make little widgets and package them up and sell them, that's a huge difference intellectually to go into information and knowledge; especially something as elusive as interpretation of the human genetic code.


Q: You divulged earlier this year that your DNA was one of the samples used in Celera's genome assembly. Why did you reveal that? It seemed to upset a few people ...
A: It does seem to have done that. But it was actually a very complicated decision. The number one thing I hear in conversation ... is that people are afraid that their genetic code will be used against them. To me, there are two types of leadership. In Vietnam (Venter served as a Navy corpsman during the Tet offensive in 1968), there were the leaders that pushed the small guy out in front to be the point man to step on the booby traps and get shot first. And there are the leaders who actually led and got people to follow them. I've never been one to push people out in front of me to get shot first.

I was a donor out of just absolute scientific curiosity. My view is, how can anybody possibly work in this field and not want to know their genetic code? Something's wrong with them! I mean what the hell are they doing? It's the ultimate dishonesty. They're advocating other people should do this. 'We're going to interpret your life but boy, stay away from mine?' I'm not shy about that. I wanted to know my own genetic code to understand my own life. I'm writing a book about that.


Q: Did you identify genes that showed you're predisposed to certain diseases? Did you go that far?
A: Yes. In fact, some of the people at Celera have shown slides at public meetings. They were showing some of my disease genes.


Q: Anything fatal?!
A: Not yet! I'm sure there's something in there. But after a while, it became more an issue of when, not if, I would disclose it. I didn't disclose it early on because [of] the media circus that was around everything we did. How many days was I on the front page of the newspapers for minor utterances? I didn't want there to be a circus and a field day around what, to me, was a very personal decision — a decision about trying to show leadership in the community, that you don't need to be afraid of this data. I didn't want Francis Collins and people turning that into a circus event.


Q: You mentioned your "disease genes." Are you taking medication based on some of these?
A: I'm taking a statin for the one gene that was publicly disclosed, an APOE e4 allele [variant] that I'm heterozygous for [carry one copy]. The homozygote [two copies] has very increased risk of cardiovascular disease. In animal models, the heterozygote has an increased risk. Most of us know something about our genetic history. You look at your parents if you have that luxury in life. My father died at age 59 from sudden cardiac death.

You don't have to be a genius to know you have some kind of genetic history there. It makes a big difference psychologically to know what you really have. To me it was an added motivation. I thought about taking statins for a long time, but it sort of pushed me over the top in thinking that I should do that. You're looking at my breakfast [bagel and cream cheese] so obviously it hasn't modified that.


Q: Earlier this year, three of the authors of the public consortium paper published a critique in the Proceedings of the National Academy of Sciences of the independent validity of Celera's sequence data. Care to respond?
A: That was just their warm-up. Maynard Olson has done what many people have told me is probably the most vile article ever published in the scientific literature, with his reminiscences of the genome project in the Journal of Molecular Biology of all places. [Olson, M.V. "The human genome project: A player's perspective." 319: 931-942 (2002).]


Q: So what was it about that critique from Eric Lander (Whitehead Institute), Bob Waterston (formerly of Washington University in St. Louis, now University of Washington) and John Sulston (Wellcome Trust Sanger Institute) that was logically flawed?
A: You can just go through the written rebuttal [Myers, E.W. et al. Proc. Natn. Acad. Sci. U.S.A. 99: 4145-4146 (2002)], but one of my biggest pet peeves in science is intellectual dishonesty and hypocrisy. The analogies I use are this quiz show, The Weakest Link, and Survivor. Basically there's a concerted effort to vote me off the island. People said they couldn't understand what possibly could be the basis of those shows, and I said you've never been on an NIH [National Institutes of Health] study session before. Of course you get rid of the strength first so the rest can compete.

So instead of having the White House event and the joint [genome] publication be a true intellectual peace treaty — I mean it was for everybody on our team — it was sort of the start of assaults that have been steadily increasing from these same three or four people. It's hard to understand. Lander is a very smart guy. He's got better mathematical training than I did and used basically mathematical sleights of hand to fool people. To me, that's intellectually dishonest because it means he knows there's something different. There's a different set of facts. But the goal is just to fool people's thinking to vote me off the island.


Q: Do you think that lingering attacks are going to damage the validation of the Human Genome Project and whether it is eventually deemed worthy of a Nobel Prize?
A: Well, what I've said jokingly is maybe some of my colleagues aren't so smart, that they can count to three! It was never my goal with this to get a prize. In fact, I probably have gotten more scientific awards per capita than almost any other scientist because of the success of this work. But nothing compares to the satisfaction of having done it. I think publishing that [human genome sequence] paper in Science, on top of the Drosophila paper, had to be a reward beyond which [anyone], unless a scientist made that level of discovery and had that impact, could ever understand.

Today in Science, we published the Anopheles genome at 270 million base pairs, and in Nature, at the same time — I think it's the first time any scientist has had the covers of Science and Nature simultaneously — published two versions of the malaria genome. I, along with Steve Hoffman [of the Naval Medical Hospital], initiated both projects, worked with Anthony Fauci [of the National Institute of Allergy and Infectious Diseases] to get them both funded and going. The genome of malaria is 30 million base pairs. We published at TIGR the first malaria chromosome, shotgun sequenced, in 1998. The malaria genome just got finished with [Wellcome Trust] Sanger and TIGR working on doing the distributed approach.

Since we published that first malaria chromosome, we sequenced the Drosophila genome and published it, the human genome, the mouse genome, and the Anopheles genome that was done in less than a year. So all those genomes were done in the time it took to do the 30-million base pair malaria genome, even using partial chromosome shotgun sequencing — just showing the value of the approach, the data. It's been validated over and over and over again. And no amount of disinformation from a few people will change the scientific validation of that.

One of the things in this Maynard Olson diatribe is how Celera left a mess of the Drosophila genome for the public effort to clean up. See if you can find a Drosophila scientist anywhere in the world to validate that statement. So what's going to be published now? The Drosophila community has closed all the gaps in the Drosophila genome.

So here's how you validate science. Now they have the complete, totally accurate Drosophila genome so they can compare back to what we had in 1999 after four months of work, and that's going to be published by people who've done that independently. And they've shown that there's less than one error [per] 19,000 base pairs in the whole genome and that the structure was right with very few errors.

I think that kind of independent validation will be done eventually with the human and the mouse genome — if you can find the copy of the human genome that the public wrote their paper on in Nature. It'd be a miracle if you can find it, number one, because it was never in GenBank and they tried to make sure it was not part of the historical record because they were doing the sleight of hand to replace it eventually. But if you can find the copy of the genome that they used for their Nature paper and compare that to the completed genome, and you find the copy that you can get from the Celera Web site or from the DVD — which I have a copy of and Science has a copy of — and you compare that to the completed one, it will tell the real scientific story of what was accomplished.



Cover co-op: Venter spoke to Bio·IT World the day that Nature and Science published the genome sequences of the malaria parasite and mosquito, respectively, featuring major contributions from his groups at TIGR and Celera. It is perhaps the first time that a group has been featured on the cover of the two rival science journals the same week. The publications were also notable for the cooperation between scientists at Celera and the public genome project.
 
Q: Is it true that Celera subsequently assembled the human sequence as Mark Adams (Celera's head of genomics) said back at the end of 2001?
A: Yes. If we made a scientific mistake — and it's an important one — it was not believing enough in our own methodology. And we had these massive debates early on, because I wanted to assemble the genome after 1X [coverage] and then 2X and [Celera head of bioinformatics] Gene [Myers] and others said no. They were almost brainwashed by the so-called Lander-Waterman equation, which had nothing to do with our new method of genome assembly using the paired end-sequences. It was the bricklayer's method of just comparing two sequences. But Gene was certain that it wouldn't assemble a 3X or 5X [sequence]. And it wasn't until we sequenced the mouse genome we stopped at 5X because there was no mouse genome data other than our own. We assembled at 5X, the same coverage we had for humans and it went together 10 times better.

And so we went back [to the human sequence], leaving out all the public data, and the human genome went together almost a hundred times better. So the ultimate irony in this is the human genome data ... was so full of errors. I think what most people don't realize is the hoodwinking that was done. They didn't sequence BACs [bacterial artificial chromosomes]. They sequenced 1,000 clones from BACs and then concatenated them together randomly to look like there were complete BAC sequences. If you read between the fine print they did say that. But it was very, very fine print. And in fact, that was the reason for shredding the data — we knew so many of their assemblies were false. But even shredding didn't break it up enough, and the data were so low quality it confounded the algorithm and degraded the assembly.

So Gene and the whole team and myself should have had far more confidence in our own ideas. We were just working with the assumption, we're trying to make a model of the genome that people would use for finding disease genes and the more data you include the better. It had nothing to do with the 'race.' The smartest thing we could do for this mythical race would've been to not use their data, because our assembly would've been so much better. And what I'm hoping is that [Michael] Hunkapiller and others will see the importance of that for validating what they've done, and they'll release those data and let it be published and compared against the genome. So, the whole genome assembly with just the Celera data is spectacular.


Q: Looking back on your career, if everything had gone right, what would have happened?

"The goal that I have is to make genomics transform medicine. That's not going to happen in academic Babel towers, talking about it." 
A: All you have to do is look at what I'm doing now to understand what my business model for Celera was. So, the model from the beginning, as I said, was to get this information to transform medicine. The databases were trying to do it. I set up a model that I think, when it gets analyzed without all the emotions, will be viewed as a breakthrough model [in which] the data can be available for free in the academic world — the raw data — because nobody knows what to do with it, quite frankly. That was the irony. I knew that. The rest of the people gave lip service to how important the data were. People wanted to patent it, but they didn't even understand what it was at the same time. Building the best tools and the best computers — that can only be done in private industry to translate this information and make it useful for people.


Q: Would you do another Celera where you could take your vision to its ultimate conclusion commercially?
A: Well, unlike many scientists that secretly have five or six companies while they're pretending to be pure academics, I candidly understand and appreciate that for things to really impact society in the U.S., there has to be a commercial component. I don't view it as an evil. I think it has the potential to be evil, but I view the hypocrisy in academia to be a bigger evil.

So right now, the science that I'm doing, because it's in an early stage, is best suited for a not-for-profit academic environment. But once I was certain that it would work and transform things on a broad consumer scale I wouldn't hesitate to work with [Microsoft Corp. chairman] Bill Gates forming an enterprise to do it.

The goal that I have is to make genomics transform medicine. That's not going to happen in academic Babel towers, talking about it. I could totally envision an information company that could ultimately be far more or as successful as Microsoft is. It's hard to imagine being more successful than Microsoft, right? I'll settle for it being as successful. Or it could be a hundred companies that do it. I don't think it really matters — my goal is to get it to happen.

Part II of the Venter interview will be published in the December issue of Bio·IT World.

PHOTOGRAPHS BY FURNALD/GRAY








For reprints and/or copyright permission, please contact  Jay Mulhern, (781) 972-1359, jmulhern@healthtech.com.