Mapping Out the Future of Genomics

IRA FLATOW: This is Science Friday. I’m Ira Flatow. In the 1990s, there was a race between two groups to create the first map of the human genome. There was the US government-backed Human Genome Project, started in 1990. And Celera Genomics, a private biotechnology company founded by biologist Craig Venter. And it began working in 1998. And Venter’s group published their results in 2001. The government officially announced a completion of its project in 2003.

Well, back in 2003, Craig Venter joins us to talk about some of his predictions for the technology. And all throughout this year, to celebrate science Friday’s 25th anniversary, we’re dipping into sci-fi archives to see how things have changed over these 25 years. And Craig Venter joins us again today to talk about how far genomics technology and costs have come down in the past decade. And these days Craig Venter is founder and CEO of the J. Craig Venter Institute in La Jolla, California.

And if you have questions you’d want to have sent our way, you can give us a call 844-724-8255. You can also tweet us @scifri. Welcome back to Science Friday.

Well, thank you Ira. And happy 25th anniversary.

IRA FLATOW: Well, thank you very, very much. How has the field and industry of genomics changed since 2003? Is it enough that you could sum it up?

CRAIG VENTER: Well, I’d say that technology has changed to a wonderful extent, both on the computer front and the DNA sequencing front. Our knowledge of the genome hasn’t changed a whole lot since 2003. But it’s about to start changing rapidly now.

IRA FLATOW: Why do you say that?

CRAIG VENTER: Well, one of the key things for understanding the genome is to get very large numbers of genomes. So we can understand out of our 6.2 billion or so letters of genetic code, understanding the less than 3% that we have difference amongst the entire human population. So with all that information, we need very large data sets to understand the differences and their significance. And that’s where the cost and the speed for sequencing has had such an immediate impact.

So two years ago, we passed the threshold that I’d set of getting below $2,000 for a human genome. And we started Human Longevity, a for-profit company, just a little over two years ago. And we’re sequencing the genome now roughly every 15 minutes versus the first one that took 9 months and cost $100 million dollars. So we can now afford to do large numbers of them and make these kind of correlations for the first time.

IRA FLATOW: And now people can send away and get a little bit of their own genome in the mail, which they couldn’t do before.

CRAIG VENTER: That’s right. There’s a lot of places that would give you partial information, but a key thing to get it done completely accurately is important for not only understanding the data but making good clinical interpretations as well.

IRA FLATOW: When you created your first human genome map, you used something called the shotgun method. How does that differ from what you’re using today or what is being used today?

CRAIG VENTER: Well, basically the shotgun method was blowing the genome apart into little tiny fragments and sequencing those. And then using the computer to reassemble that into the genome sequence. So basically, since 1995 when we did the first genome and certainly after using that for the human genome, it’s literally the only way that genomes are sequenced.

The only difference between what we do today and what we did in 2001, in 2001 we had to sequence the genome de novo and actually assemble it back into a sequence. Today, what makes the reads faster and cheaper are making very short reads, so only around 100 letters of genetic code. And those aren’t adequate for de novo assembly of the genome.

So basically, everybody cheats. And they line up those fragments either on my reference genome or a composite one that NCBI has put together. And you do it sort of just by matching the fragments versus truly assembling the genome. So we’re cheating until technology allows us to actually do routinely what we could do 15 years ago.

IRA FLATOW: Interesting. When you were on the show back in 2003, this was after you left. You were fired from Celera, the company that you founded. And this is what you said about the split.

CRAIG VENTER: There was a fundamental disagreement on the future directions. I was apparently the only one that believed that genomics was the future of what would change in preventive medicine, changing how we practice medicine, even the cost of medicine. The heads of the parent corporation of Celera thought that going into making pharmaceuticals was the only way to go in terms of getting return on their money. It was just a disagreement in direction. And I let it be known that I would probably move on. And they helped make that happen.

IRA FLATOW: And it was fortuitous that you left.

CRAIG VENTER: I think it was. I’m still here but Celera is not.

IRA FLATOW: There you have it.

[LAUGHTER]

CRAIG VENTER: So I think that’s called the tale of the tape or something like that.

IRA FLATOW: Sounds like it. It’s almost the Steve Jobs story. You and he had the same. You left, came back, and started your own company.

CRAIG VENTER: Well, it’s creative minds working at the edge. It’s not always easy to get investors to think in that same out of the box fashion. So it’s a challenge to do so. And we’re pushing the limits now at Human Longevity. But it’s exciting and we’re starting to make so many unique discoveries about ourselves that it’s going to be a very, very different world trying to understand our DNA software.

IRA FLATOW: I’ll bet. And back then in 2003, you had a prediction about how widespread personalized genome maps would be. Let’s hear what you said back then.

You have written that you think that within the near future, everyone will be able to walk into their doctor’s office, maybe give them a sample of some part of their body, and get their genome back for a fee. How close are we to that?

CRAIG VENTER: Well, I predicted that that would happen in 10 years. Obviously, we need some major technology changes. When you think of the public genome effort requiring $3 to $5 billion to get one genome, obviously we have a long way to go if we’re going to get everybody’s genome that way. That doesn’t make it very feasible. Even the Celera effort cost on the order of $100 million and took 9 months. I mean, we’ve made it an international goal to get the $1,000 human genome, which we think is the range the technology has to drive things to to make it so that lots of people can afford, insurance companies will afford, to have a genome sequence for each person.

This is something that only has to be done once in your life. It’s not a diagnostic test you keep going in to do over and over again. So I predicted that will start to happen with children born in the hospital, that before they leave the hospital their parents will have the option of having that information.

IRA FLATOW: That hasn’t quite happened quite yet, but do you still believe in that prediction?

CRAIG VENTER: Well, in fact, a few things I said there were wrong. But the others were– it’s now 13 years instead of 10. And this year we’re actually rolling out those efforts to do newborn genome sequencing. And we’ve been doing major diagnostic sequencing on children with undiagnosed diseases in hospitals. And we’re much closer to making that a standard for health care.

One of the things I said that was absolutely wrong is I predicted that we’d only need to get our genome sequenced once. And one of the findings at Human Longevity as we sort through the genetic code using our machine learning tools, is we can actually predict your age from your genome sequence. I wouldn’t have guessed that 13 years ago. In fact, I clearly didn’t. So, in fact, the notion that you only get your genome sequenced once unless you have cancer is going to probably end up being not correct.

We’re finding specific changes in the genome as we age. As I said, to the point where we can predict your age relatively accurately straight from your genetic code. So understanding those changes could be the key to understanding how we age and why we age.

IRA FLATOW: Can you predict my death then, from the genetic code?

CRAIG VENTER: So that’s something of hot discussion right now. And technically, not down to the hour and the day, but probably within a few years, we think that’s possible. It’s not clear how many people want to know that. There may be other people that want to know that about you, but you may not want to know.

IRA FLATOW: I mean, insurance companies are going to want to know that probably.

CRAIG VENTER: Well, certainly it’s interesting. We’re having a lot of discussions within Human Longevity and life insurance companies. And one thing you have in common with your life insurance company is they want you to live a long time and you want to live a long time. You’re betting against yourself and they’re betting for you. But they’re both going to be interested in that information.

I think the challenge will be if the insurance companies have that data and you don’t, you’re at a disadvantage. And if you have that data and they don’t, they’re at a disadvantage. So I think it’s going to be an interesting period to make sure that there’s sort of equilibrium, that you can’t fool the insurance system by having an idea when you’re going to die or what you’re going to die from, or what you’re not going to die from.

Because that could throw the economics of the exchange into wild chaos. So I think it’s going to be important for both you and your insurance company to have that information, because what we’re finding is the information can be predictive. And you can do something about those predictions to change the outcomes.

IRA FLATOW: How accurate are the predictions, 50%, 75%, 100%?

CRAIG VENTER: Well, it depends which ones. I think one way that I think it’s important to look at just straight genomic data is in statistical probabilities, not in yes/no answers. And so, we can predict your age with an accuracy maybe plus or minus 5 years right now. It’s not a precise element, but should get more precise with time.

Predicting when people would die, it’s interesting because actually there’s genetic changes that we see that actually cause people to die early. And those predictions are certainly within five or 10 years right now of accurate predictions. But what they give us is a chance to change those outcomes. These become the new targets for pharmaceuticals and actions, so you don’t die from these things precoded into your genome.

IRA FLATOW: Yeah. Unless you’re texting while driving, and there’s a whole different ball game. Well, one of the biggest questions that comes up all the time, and it came up way back in 2003, is who owns the genes? Can a company patent them or patent it? And this is what you said back then.

CRAIG VENTER: What was used by our early critics was saying that we were going to patent the human genome, even though they knew that was false. So it’s been a complicated issue mixed with politics. But as somebody who is now in his mid-50s, I’m delighted that some companies are patenting genes associated with potential new therapeutics.

What I’d like to see is less of it on a speculative basis and more of it when there was a direct link to making a drug. I think if that was the case, all of us would be a lot happier with the patent office.

IRA FLATOW: If just joining us, talking with Craig Venter on Science Friday from PRI, Public Radio International. In 2013, the courts ruled that Myriad Genetics could not patent the BRCA1 and BRCA2 genes that are related to breast and ovarian cancer. Should companies have to pay royalties on a gene, Craig?

CRAIG VENTER: Well, I don’t think anybody should in terms of diagnostics. And in fact, doing genome sequencing would have put that era to an end anyway, because we can sequence your entire genome and have your sequence of every gene for far less than it costs to just get your BRCA1 gene tested. And it’s not something that could have been covered by patents, because we couldn’t be told well, we can’t look at that piece of the genome when we’re sequencing it.

But I think we were all delighted with the Supreme Court ruling. And that you cannot patent naturally occurring DNA. And I think that’s really opened up the diagnostic and the genomic arena. So we’re all concerned the genome would get balkanized like it looked like was happening with BRCA1 and BRCA2.

IRA FLATOW: Your latest work is on synthesizing DNA. And you actually created some synthetic DNA. Put it in a cell recently. Remember, we talked about it. How close are you to building a cell from the ground up, from the cell membrane, the organelles, everything else in nature, and putting in the genetic material? Or is that not something that is on your radar?

CRAIG VENTER: It’s an interesting idea. People keep talking about life from scratch. And it’s like when people talk about baking a cake from scratch. It doesn’t mean they’re going to go out and make the flour from carbon and CO2 and water molecules. It means not using the cake mix.

Life from scratch has lots of different meanings depending on who’s saying it. You can’t have life without our genetic software. So just putting proteins together in a lipid bag, even if you get metabolism from it as some people have, that’s not life. It can’t sustain itself. The proteins die off.

And so without genetic software, you cannot have life, and certainly self-replicating life. So we focus totally on the building and design of that software. And just using the existing cells to boot up the software to create totally new life forms. At some stage, it would certainly be interesting if we could put the DNA together with the right lipid bubbles and maybe a starter reaction of ribosomes and things. So you can start making proteins and getting cells that way. Because the more we can break things down to basic components, the more we will have to understand them to get it to work.

And that’s what the big finding was with the cell we just spent the last five years trying to design. We tried to design a minimal cell with just the genes necessary for life. And it turns out the last 20 years of science has been misleading, because we all just sort of subtracted one genome from another and got down to a set of null genes.

IRA FLATOW: I have to interrupt you there. We’ll come back and talk about that. We did talk about how very few genes you needed to keep that cell going. Talking with Craig Venter, founder and CEO of the Craig Venter Institute in La Jolla. We’ll be right back after this break. Stay with us. This is Science Friday. I’m Ira Flatow. My guest is Craig Venter. He’s a biologist, and founder and CEO of J. Craig Venter Institute in La Jolla, California.

And Greg, I don’t mean to try to beat a dead horse. I just want to see if I understand correctly. Do you think it’s too big a hot button issue to make a living cell from scratch, like you say baking a cake? To try that, it’s just not worth the heartache you might have to go through to create?

CRAIG VENTER: Well, no, I don’t think it’s a hot button issue at all. I think various researchers are trying to replicate what they might have as origins of life. So Jack Szostak at Harvard has been making small lipid soap bubbles and putting self-replicating RNA molecules in them. And the bubbles sort of separate into daughter bubbles. And the RNA molecules replicate themselves.

So showing that those could be early origins of life. We’re dealing with things several billions of years later, even with the most minimal cell. So it’s not like we’re recapitulating the first creation of life with these basic elements. I just don’t think it’s that relevant of a question given that everything derives from the DNA software itself. And I’m sure in a few years, we will have universal cells that can read the ranges of types of genetic code to boot up all kinds of new cells.

So I think once you have computers, you don’t go back to an abacus.

IRA FLATOW: I see. That’s sort of going in reverse. If you get everything you need out of the DNA, why go back and try to make the cell? Let’s go to the phones. Let’s go to Cleveland Heights, Ohio, Barry. Hi, welcome to Science Friday.

BARRY: Hi, thank you. So I’m thinking about all of this technology we have with respect to acquiring information about the human genome. And we’ve had a lot of this information before. And I’m wondering how maybe Mr. Venter’s ideas are regarding this idea of eugenics and how that’s been used in the past. And how it might be used differently now that we have this technology. And also about this idea of the CRISPR technology. And maybe he can elucidate a little bit about what that is and how it might be related to the topic.

IRA FLATOW: OK, thanks.

CRAIG VENTER: Well, Barry both interesting questions. So eugenics came to a peak in, I think, the 1930s, long before we had DNA sequence or even knowledge that DNA was the genetic material. But people wanted to do different types of selection or thinking that they were going to eliminate populations that had criminal or bizarre behavior.

So it’s something that has been of concern to everybody since the start of genomics. And I think thus far we’ve been very good to keep that out of it. But as we understand more and more of ourselves and how our personalities and behaviors and brain structures are programmed from DNA, we’re going to have to be careful we don’t start down some slippery slope.

CRISPRs are a DNA editing tool, and they’re a fantastic discovery, a fantastic tool for making small changes in the genetic code. There’s many people thinking and talking about doing this with the human genome. I personally think this is very dangerous to do until we have a far more complete understanding of the human genome.

We were just talking about a minimal bacterial cell where we don’t understand a third of the genes, with less than 500 genes in a cell. We know less than 1% of real information about the human genome, human genes. And it would be a complete human experimentation to start making changes in genes to see what the outcome of those would be in living humans.

So I think we need to be extremely cautious and at least wait till the consensus of our knowledge is sufficient that we wouldn’t end up doing grave harm to people by making changes that we don’t know what they really mean.

IRA FLATOW: As you know, there are diseases that are based on simple genetic mutations. And some of the genes that could be replaced by CRISPR. And you’re saying, be careful in moving forward in that direction.

CRAIG VENTER: Well, if you look at other species where people have studied things, so for example, if you’re looking at a structural protein in the wing of a fruit fly. And say, we understand the damage in that. We’re going to correct that error in that protein in that gene. Turns out that same protein and gene early on in the development of the fruit fly play a key developmental role. And it’s just later in the fruit fly’s life it ends up a structural protein in the wing.

So assuming that every gene only has a single function is probably the fastest road to hell in genomics. So we need to understand all the functions of genes and how changing them will alter other genes, other biological functions in humans. We’re not simple, linear creatures, and we have to understand how changing one note changes the entire orchestra.

IRA FLATOW: Yeah, because we found that we have very, very much less genetic material than we thought we did. And that it must all be working together in different ways, not as simply as we thought.

CRAIG VENTER: Yes, the gene-centric view of the world we’re hoping will die away. And we’ll talk about the genome-centric view of how all these parts work together in unison to create our cells and our bodies.

IRA FLATOW: Well, in the few minutes I have left, a minute or two, you want to make any predictions for 13 years from now when we have you back?

CRAIG VENTER: Well, I hope I’m around to come back in 13 years.

IRA FLATOW: Me too.

CRAIG VENTER: But from my genome, I’ll cautiously make that prediction. I think in 13 years genomics will be a very key, fundamental part of the practice of medicine. Completely in cancer, you won’t be able to do anything in cancer without understanding the genetic code. And I think it will rapidly move into other areas as we get this. 13 years from now, we’re going to have tens if not hundreds of millions of genomes and databases that will give us a very different view of humanity than we have today.

IRA FLATOW: We’ll be crowdsourcing big, big data. Genome will be big data.

CRAIG VENTER: It’s already big data. So just the 26,000 genomes we’ve sequenced at HLI in the last two years reference 3.4 petabytes of data. Indexing the entire internet in 2014 was only around 12 to 14 petabytes. So we’ve already surpassed the big data definition in genomics.

IRA FLATOW: And one last question, what big project are you working on that we’re going to be hearing about?

CRAIG VENTER: Well, the key thing we’re doing is integrating, actually measuring, clinical tests with MRI and some of the other measurements we do, linking that to the human genome to greatly increase the understanding. And being able to do predictive, preventive medicine. We’re discovering cancers very early at stage zero and stage one when they’re completely curable, instead of at stage four, where maybe you’ve got a 50/50 chance of survival.

IRA FLATOW: And we’ll see the fruits of that results soon?

CRAIG VENTER: I think every few months that go on, we’re going to see more and more results of it. I think by the end of this year, there will be major new programs that we have that make it much more cost effective for people to get cancer, genomic information to help save their lives.

IRA FLATOW: We’ll be looking forward to it, Dr. Venter. Thank you for taking time to be with us today.

CRAIG VENTER: Ira, it’s always great to talk to you. So again, happy anniversary.

IRA FLATOW: Thank you, and good luck in your work. J. Craig Venter is founder and CEO of the Venter Institute in the La Jolla, California.

Copyright © 2016 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of ScienceFriday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies.