J. Craig Venter is the pioneering cartographer of the human genome, the sequence of which he and other scientists mapped in 2000. The WorldPost recently spoke with this modern Prometheus about the promises and perils of being able to read, write and edit the human genome.
You have said that humankind is entering a “new phase of evolution” -- from natural selection to intelligent direction. Why is this so, and what does it mean?
Biological evolution has taken three and a half or four billion years to get us where we are. Social evolution has been much faster. Now that we can read and write the genetic code, put it in digital form and translate it back into synthesized life, it will be possible to speed up biological evolution to the pace of social evolution.
On a theoretical basis, that gives us control over biological design. We can write DNA software, boot it up to a converter and create unlimited variations on biological life.
This year is the fifth anniversary of when my team produced the first synthetic cell. To do that, we took the ones and zeroes in the computer, rewrote the genetic code from four bottles of chemicals and booted that up to get a self-replicating cell. That means we now have the power to start controlling evolution.
We’re doing this now in cells that can change manufacturing and create a new industrial revolution by creating synthetic food, chemicals and even building materials. Ultimately, as we begin to better understand our own genetic code, we can edit the human genome -- as some Chinese scientists disturbingly did earlier this year.
So we have the power to do it. But we clearly do not have the wisdom to do it or the knowledge to do it in a safe fashion. That is why many of us involved in the science have suggested a moratorium on any human changes until we understand the full consequences of our interventions.
In the end, however, it is inevitable that we will not be able to control ourselves. Using knowledge to eliminate horrific diseases from the population is going to be an overwhelming temptation. The flip side of eliminating disease will also be irresistible because we have learned now how to improve intelligence and how to improve athletic abilities -- in short, how to make better people.
Your concern with gene editing, then, is not that it shouldn’t be done, but that it shouldn’t be done until we have adequate knowledge of the consequences. What’s going to make the difference?
The difference will be when we attain more complete knowledge, which we just simply don’t have.
Let’s take the example of the fruit fly. If it has a defect in its wing, we can find out today what structural protein has caused that defect and fix it. But it turns out that same gene that leads to that protein structure early on totally controls key development of the body as well as the wing. Only in its final stage does it become a structural protein. So if we intervened and thought we were just fixing the wing, we could be deforming the whole body of the fly.
If we assume we know what a gene that we’re changing does, but it actually changes overall development or changes some other process we did not have proper understanding of, that’s called “experimentation.” And that is a pretty dangerous thing to do with humans.
It is totally a false notion that you can de-identify or make anonymous the human genome.
If we try to author major changes in humans, more problems that are unforeseen will be inevitably created. There’s lots of science fiction movies about that, obviously, from Frankenstein on.
We simply have to have more thorough knowledge so that we can make changes in a beneficial way for society and in a safe, effective way for those individuals.
You are the pioneering cartographer of the human genome. How much do we know? What percentage of the functions of genes do we know today?
The cell that we’ve designed in the computer has the smallest genome of any self-replicating organism. In this case, 10 percent of the genes, or on the order of about 50 genes in that organism, are of unknown function. All we know is that if certain genes are not present, you can’t get a living cell.
The human genome is almost the flip side. I would say that we only know well the functions of, maybe, 10 percent of our genome. We know a lot about a little bit; we know far less about a lot more. We don’t know most of the real functions of most of the genes. A big percentage of that can potentially come in the next decade as we scale up to get huge numbers and use novel computing to gain a deeper understanding.
The obstacle, then, is big data analysis computing power? You have the basic information. The question is how do you analyze all the functions?
Deciphering my own genome cost $100 million to do. It took nine months and a huge factory full of machines. That is obviously not a very replicable event. But now technology has become a whole lot cheaper and faster, though it is not as accurate as we had fifteen years ago.
Now the cost is down to about $1,500 to decode a person’s genome. So for the first time now, we can start scaling up to get very large numbers. Fifteen years ago, matching 10,000 genomes was clearly something off in the distant future. At $100 million each, it wasn’t very likely. Now we’re trying to get a million genomes under our belt, so to speak. But the species is approaching 8 billion people. That’s still a tiny fraction of humanity.
So, we have a long way to go. To compute and understand all the information that we are collecting so we can understand it creates a huge data problem. It is a multi-factorial problem. The notion that people have had that there’s one gene for diabetes or cancer and all you have to do is find it and fix it is going to disappear from science pretty quickly as we understand multi-component, multi-factorial bases of human function and disease.
About half our genome codes for our brain. That’s not three genes that code for the brain, but about 10,000 and some odd combinations that we don’t know about.
One thing we are doing now is trying to predict a person’s facial features from their genetic code. To do that, we’re taking about 30,000 measurements of the face and 3-D photographs in order to discover the contributing components across the genome.
So I think that we’ve had these way over-simplistic notions of genetics and genomics based on linear thinking. Fifteen years ago, a lot of scientists were hoping to find only 300,000 genes so there’d be one for each trait. Unfortunately, that’s not how our biology works, or we’d be pretty simple organisms.
How are you discovering the genes that determine a person’s facial features?
The way it works in reality is that your genes determine your face, so it’s not a wild stretch of the imagination that it might be doable, right? We all look a little bit different based on the small differences in our genetic code.
We have a series of cameras that snap a 3-D photograph of faces and take about 30,000 unique measurements -- the distance between your eyes, for example, and other physical parameters. We then look into the genome for those 30,000 measurements to see if we can find parts of the genetic code that clearly determine that factor.
Obviously, there’s a lot of variation across the human species, so it’s not a simple algorithm. I'm less confident we will be able to take your genome sequence to predict your voice, though, but we’ll get approximations of it. Perfect pitch is genetic. Cadence is genetic. But there are a lot of other things that go into how we sound.
By way of analogy, the U.S. government has defended its NSA spying programs by saying that metadata enabled them to look at patterns but did not intrude on the anonymity of the individual since no one’s name was attached to that data. But what you are saying is that, with the metadata of the genome, we can go back in and determine individual characteristics.
It is totally a false notion that you can de-identify or make anonymous the human genome. We need to change the laws regarding donation of genome information for research because otherwise people are under false notion when they are contributing their samples thinking that they can be completely anonymized.
If we can generate a photo of you and a pretty accurate description of you from your genetic code, it’s going to be a lot harder to be anonymous. That is especially true now with all the photo identification software that’s out there. For most things, you only need two data points to make a link to an identity.
You can’t get your phone number from the genome, but with the Internet and a few simple tools, such as Google Faces, it’s not hard to make links, unless somebody’s totally off the grid somewhere and has never appeared on the Internet.
If we have a science-illiterate society, maybe they shouldn’t be trusted with this complex information that can predict human traits accurately.
So, yes, we need to be much more careful about it. At our labs, we are creating a highly secure database, and you will not be able to scroll through it and find your genome and run an algorithm. It’s amazing that this silly notion that you can de-identify a person once you have their genomic information has lasted as long as it has. We need to be very careful in telling people that they can be anonymous donors.
Last year, a group published a paper linking to my family members and finding a lot of information because my genome is on the Internet. My advice to myself 15 years ago, if I could do it over and to anybody else, is to know a whole lot more before you make the decision to make your genome public.
I would not advise a 20-year-old JCraig Venter today to do that because we don’t know how our society is going to deal with that information. Is it going to be a positive use to determine the common thread of disease, or are we going to go back to the 1930s when the Cold Spring Harbor Laboratory was the home of the eugenics movement? Let’s remember that we live a society in the U.S. where over half the population doesn’t believe in evolution. If we have a science-illiterate society, maybe they shouldn’t be trusted with this complex information that can predict human traits accurately.
Is a voluntary moratorium sufficient to control such consequential technology?
We in the U.S. certainly can’t control what the rest of the world does. We can make it so those results don’t get published in top journals or so that no one is rewarded for doing human experimentation. That’s been the status and approach we have taken to this ever since even all the Nazi experiments.
Even though that information might have been beneficial, basically all that data was banned and not allowed to enter into modern science to the extent that it could be controlled. If people are looking for respectability and approval from their peers, which is one of the goals most scientists have (versus trying to just make a lot of money), there is some ability to perhaps control the process. But in our world, control has to be on a state-by-state basis. I can’t imagine that the Chinese government wants to create a number of new burdens in society by having human experimentation go awry.
You are also working on what you call “humanizing pigs.” What is that project about?
My company, Synthetic Genomics, has a major program that aims to change the pig genome so that pig organs -- hearts, lungs, kidneys -- can be used for organ transplantation into humans.
About a million people die in the U.S. each year due to the lack of organs for transplantation. Even with all the people dying in car accidents, we have not caught on to the notion that we should make our organs available for helping other people live.
Pig organs, particularly the heart and lungs, are about the same size as human organs, and if we can change a number of genes so they have human genes and human proteins, that will allow them to be transplanted without massive rejection. Hearts can now last a little over a year in these transplants. Lungs are the most difficult tissue. They only have lasted very short periods of time. My company’s goal is to make new cells that can then be deployed in nuclear transplantation to create pig embryos and therefore new pigs with the humanized tissues. It’s a very rigorous process and we have a massive number of people working on it. It’s not trivial to do but it is such a key medical need that we are giving it a go.
How far off in the future is that? 15 years?
Well, we hope it’s not that far off. We have to perfect the process. We are doing the design and trying to change things. All the processes of rejection aren't totally understood. There is a very rapid phase, there are intermediate phases and there are long-term phases. But we are optimistic that it will be doable. We are counting on less than 10 years. We are pushing for five.
Finally, your most ambitious project is to extend human longevity. Can we live to be 150 years old?
Our goal is not to live so many more years but to live a healthy life span. We need to think carefully about societal consequences if of all of a sudden we all start living to 125. We are overusing the planet’s resources now as it is.
If we are going to live longer, we need to learn how to construct a reusable, recyclable and renewable lifestyle versus just a consumption one. We’re trying to help that with biology by capturing carbon dioxide and converting it back into chemicals or even energy. We have to change what we do before just making people live longer.
The goal is to it change people’s quality of life so we can live whatever a normal life span is without massive disease. We all know people with cancer, with heart disease -- and the consequences for their families and the society around them. If that can be postponed or eliminated, that’s a huge impact on humanity.
This interview is part of the WorldPost Series on Exponential Technology.
Also on WorldPost: