George Church and Ramez Naam on the limitations of evolution, the power of matchmaking, and why we should send single-cell computers into deep space.
George Church is a geneticist at Harvard Medical School, but that title undersells him. He is very much an engineer, having developed technologies for sequencing DNA, editing genes, and manipulating stem cells. He’s also a beguiling prophet of radical biological change who urges caution even as he reminds people how tantalizing it all might be.
Where is all this stuff heading? To draw Church out on such questions, NEO.LIFE turned to Ramez Naam, another creative thinker who defies easy categorization.
Naam is the chair of energy and environment at Singularity University and author of More Than Human: Embracing the Promise of Biological Enhancement, The Infinite Resource: The Power of Ideas on a Finite Planet, and the much-admired science-fiction trilogy Nexus. So you can see why we figured he and Church would have a lot to talk about.
Their conversation has been edited for length and clarity.
Ramez Naam: George, you’re a pioneer in genome sequencing and in thinking about editing the genome. And you have been less knee-jerk than most people in rejecting some ideas that people find pretty far out. When news came from China last year that a scientist had edited the embryos of two girls in an attempt to make them resistant to HIV, there was a wave of condemnation. And you said, “Hold on, let’s take a balanced view.” Where does that instinct come from?
George Church: Well, it probably comes from doing technology development, including some technology development that’s made its way into clinical practice, and seeing the successes and failures in the long history of gene therapy. One thing is to ask what the long-term consequence is going to be: Is it actually likely these kids are going to die, as some people did in the early days of gene therapy?1
1A paper published in Nature Medicine in June showed that the HIV resistance gene that scientist He Jiankui tried to edit into the embryos also appears to raise the risk of death from the flu and West Nile virus. The study was based on an analysis of 400,000 genomes in the United Kingdom’s Biobank. Asked for comment on the study, Church pointed out that while influenza is prevalent in the UK, HIV is a bigger risk in other places. Assuming that edits for HIV resistance could be made accurately, the tradeoff might be worthwhile for some people. “Many new therapies, even those extensively tested in animals, face similar questions,” Church said.
My particular experience has been unusual in that I’ve seen things change extremely rapidly. The other part of what’s unbalanced in the discussion [of genetics research] is that sometimes people say, “Oh, this stuff is so far off and so impractical, so unlikely, we don’t need to really take it seriously.” That doesn’t prepare us well for a sudden change in technology. For example, bringing down the cost of sequencing from $3 billion for a poor, non-clinical genome, to nearly zero dollars now for a high-quality diploid genome2, was supposed to take six decades, and instead it took on the order of six to eight years.
2 A diploid genome includes both sets of chromosomes. Early sequencing projects generated composite genomes out of haploid sequences—just one set of chromosomes—from multiple donors.
You need to discuss the scenarios because a false positive—where you talk about a scenario that never happens—isn’t so bad. But a false negative—where you miss a revolution and then end up being reactive rather than proactive—is worse.
Naam: You mentioned an almost-free, high-quality diploid human genome.
Naam: What’s the implication of that?
Church: Well, to know the full implication, you’d need to know when society is going to adopt it. We had amazing computer infrastructure, including an internet, in the late ’80s, but not that many citizens were using it. Mostly computer geeks. And that transitioned radically when the first Mosaic browser and standards for HTML came out. I think the same thing is going to be true here. We’ve got the infrastructure in place to sequence everybody’s genome on the planet, but there’s got to be a social revolution and maybe some marketing. Just the right combination that will cause us to go past the tipping point.
I think it could happen any minute now, and it could immediately save trillions of dollars just with Mendelian diseases alone. We don’t even have to discover the basis of complex diseases and all the rest that people think is far off. Just getting the wheels turning on Mendelian diseases and getting that out to everybody equitably could completely change poverty—in industrialized nations as well.
Naam: By Mendelian diseases, you mean diseases that are caused by just one gene?
Church: It could be one or two. Simple diseases that are highly predictive and very serious. They don’t have to be curable, but you can fix them.
This has been proven with matchmaking, which is very low-cost and doesn’t require FDA approval. That’s a radical but proven and very inexpensive approach to eliminating some of those serious diseases, often heart-rending family situations.
“Gattaca meets Romeo and Juliet, with gene-crossed lovers who want to be together despite their genetic incompatibility.”
Naam: Just by letting people know what they carry and what their potential mate carries, we can avoid a lot of those?
Church: That’s the way it’s currently done. But you could do it without anybody knowing that they’re carriers. We’re all carriers of something. So what you do is, you just get a list of a large number of people who are geographically convenient, in places that you go, who are compatible with you in age, and interest, and genes. That way, you never have to have the heart-rending problem of saying, “Here are two people who are in love” and then they learn they’re incompatible, and then they split up. Everybody knows they split up because they’re both carriers. That’s unnecessary emotional harm and stigmatization.
Naam: I can see the movie script writing itself for this. It would be a dystopian state, where the state does this matchmaking service. It’d be Gattaca meets Romeo and Juliet, with gene-crossed lovers who want to be together despite their genetic incompatibility.
Church: No, I think the thing is, do it as early as possible so that anybody you’re thinking seriously about even dating is already on your list. You’re not excluding that many people. It’s like 5% of the possible dates you’d be excluding. It’s not a big—
Naam: Yeah. I mention that movie sort of tongue-in-cheek because society has this tendency to pick not-super-plausible scenarios of harm or damage for new technologies like this and fixate on them, rather than the upsides. Do you see any way to fix that? Or is that getting better?
Church: Well, I actually think it’s not necessarily all bad. Again, the consequences of having a false positive, where we get all worked up about something that doesn’t materialize, like Y2K for example, that’s not so bad. It’s much better than having it sneak up on us, and we just weren’t ready for it. There are a few internet things that we just weren’t quite expecting. I think it’s better to have the negative scenarios out there. It’s all about balance.
Naam: You’ve compared the Chinese kids who had been gene edited to Louise Brown, the first baby born through in-vitro fertilization. We used to say kids born through IVF were “test tube babies.”
Naam: Do you think 20, 30, 40 years from now it will be the same way with gene-edited human babies? That it’ll just have become normalized?
Church: It’s quite possible. But there are alternatives. One that we have talked about already is matchmaking. That solves a lot of things. If you had that working, you wouldn’t need gene therapies in adults or embryos. Or male or female infertility, if it can’t be fixed by in-vitro fertilization—and there are quite a few that can’t—then by definition [gene edits on those people] would be restoring the germline to an ancestral, healthy, functional state. I think that’s an example of something which would be considered as medical as in-vitro fertilization and might be accepted.
Mitochondrial gene editing, or gene transfer, also fits in this category as something that we already accept in some countries. I think, in a certain sense, it’s already happening, and it will be the new normal.
Naam: What about people who want to augment? What if somebody wants to have a child and says, “I am heavy and my spouse is heavy, I want them to be a good-looking, normal-weight person.” They will justify it as lowering the risk of diabetes, but really it’s an aesthetic thing; it’s an enhancement. People will want to do that, right?
Church: Yeah. I think that the people that have a particular disease are in the best position to determine whether their children can handle it or not. People who’ve never had it in their family are not in a good position to judge. I don’t think lack of obesity, necessarily, is going to be considered by everybody to be an enhancement. But there are examples of enhancement that we already have in most industrialized nations, like the immunity to 20 different infectious diseases. We’re differentiated from our ancestors. Our ancestors lived in mortal fear of these diseases. We don’t.
We’re also augmented in non-biological ways that have biological impact. For example, if you said, “Well, I’m going to engineer my muscles so I can run really fast.” How is that different from hopping in an Uber?
Naam: It’s still not as fast as Uber.
Church: Right. Or a jet. Or a rocket. So these days, the need for physical augmentation, biologically, is less than it used to be. We still want to augment, but it’s usually through physics and chemistry. I think that there’s a strange exceptionalism where we can get augmented one way but not another. We can get augmented as an adult, but not as a baby. There’s a lot of lip service to preventative medicine, which could be considered augmentation. It’s hard to actually get preventative medicine going if it’s powerful medicine. If it’s something simple like eating something different or walking a few extra steps, that easy. But if it’s longevity, then you have to frame it as reversing diseases of aging.
Naam: I’ll offer an intellectual augmentation: literacy. Learning to read at an early age rewires the brain biologically.
Church: Perfect example. Yeah.
Naam: But no one thinks of it because it’s baseline now. Are you saying that we should be comfortable with offering some of these augmentations in the embryo?
Church: I think it’s case by case. My guess is a lot of the augmentation is going to happen in adults first. One reason is, there’s only 100 million babies born each year worldwide, and there’s 7.5 billion adults, so it’s a bigger market. Also, some of the augmentations will be intrinsically easier to do in adults. Our aging society is going to get us hyper aware of cognitive decline, so we’ll want to compensate with cognitive enhancement. It’s going to be very hard to resist that possibility. Even though we say that intelligence is incredibly complicated, well, there are plenty of examples of one or two or three genes causing great enhancements of one or many cognitive abilities. It’s been shown in animals. Many of these things are candidates for gene therapy in people who have the first symptoms of Alzheimer’s or just have some risk for it. There are gene therapies in the pipeline today that are going in that direction, reducing the probability of cognitive decline in Alzheimer’s and other sources of decline. Some of those could result in cognitive enhancement if used by someone who’s not in decline.
“I would be one of the ones that you would get rid of if you were trying to squeeze the bell curve down.”
Naam: Would that would be better for society, if we had more people who were smarter?
Church: We need to be cautious about everything that’s new, even some things that sound like Mom and apple pie. We have to ask, “Are we maintaining the neurodiversity we need?” For example, autistics have historically been described as intellectually disabled. But very often, if they can become high-functioning, they’re the opposite. They actually contribute things to society that nobody else sees. Same thing could happen with a whole variety of neurodiversity.
Naam: You have narcolepsy, right?
Church: Yeah. I’m narcoleptic, and when I was young, dyslexic. I think I have little bits of OCD, as many of my colleagues do.
Naam: So you’d consider yourself neurodiverse?
Church: I would think that I would be one of the ones that you would get rid of if you were trying to squeeze the bell curve down.
Naam: Do you consider any of your neurodiversity a contributing factor for your success?
Church: My gut tells me that I have benefited, that it’s been a net positive. To some extent, just being different on any axis, it doesn’t have to be something that anybody would recognize as being an advantage. Just being different at all from the middle of the bell curve gives you an advantage in a part of society that cherishes innovation and out-of-the-box thinking. That’s a part that I managed to find. But I’m sure there are many professions, and different eras, and maybe even different geographies where I would be dead or at least penniless.
Naam: It’s interesting because in the tech world now, there are all these articles about people in Silicon Valley micro-dosing psychedelics or whatnot, and it almost reads like these people are trying to make themselves temporarily more neurodiverse.
Church: I agree. They were born in the middle of the bell curve, and they’re trying to escape.
Naam: Could I do that via genetic methods? I read long ago that one of the alleles3 that has the highest correlation with changes in IQ has an association with higher risk of schizophrenia. What if I want to make myself smarter in some way that might come with other side effects? If I’m an adult, should society allow me to do so?
3 A variation in the letters of a gene.
Church: Well, “shoulds” are difficult. The “should” is: Should society allow you to do that in order for society to achieve a particular goal like having better books? Or better technologies? I think society could tolerate at least a few more out-of-the-box thinkers. I don’t know if we have the ideal number. So anyway, it’s really a matter of, “Are you endangering yourself to such an extent that you’ve become a burden on society, where government has to pay for your medical expenses for the rest of your life?” If that results in you becoming a quadriplegic, society probably doesn’t want that to happen to you, reasonably. The same thing could happen with becoming neurodiverse.
I think that something you can turn on and off has a particular attraction over something that’s permanent, something that you did in the germline. Wouldn’t it be kind of cool if you got one of my disabilities or autism and could be able to turn it off when you need to be sociable or need to relax or something? Then crank it all the way up when you want to be creative, or have a deadline or something?
Naam: In the field of brain-computer interfaces there are people who talk about using transcranial magnetic stimulation to temporarily induce savant states. I’m not sure anybody really knows if it works or not, but it’s kind of along those lines.
Church: I tend to think it’s wishful thinking that there’s something you could eat or drink or some convenient thing to put on your head, whether it’s infrared or magnetic. When we talk about gene therapy, we would literally have complete access to the whole biological spectrum. Maybe not today, but you could see that’s where it’s headed. All the shortcuts and simple therapies we think of today are going to seem fairly pathetic.
Naam: When we talk about the whole biological spectrum, you’re not just talking about genes or alleles that exist in humans today. You’re talking about going beyond that, like genes from other species, or genes that just don’t exist at all, potentially.
Church: That’s right. We need to take that sort of possibility very seriously.
Naam: You’ve said that the human future in space depends upon editing our genes.
Church: It may not even be considered augmentation. If we’re desperately sick due to radiation and low gravity over long periods of time, then that would be considered a medical emergency. It might be acceptable. I do think that there are new challenges that we’re already facing in cities that our ancestors weren’t fully evolved for, but we’ll definitely be facing it in longer-term colonization efforts in space.
Naam: How might we modify ourselves to thrive in space?
Church: Radiation resistance is actually quite well understood. There are some extraordinarily radiation-resistant biological systems that we could move over into human cells, at a minimum. There are problems with osteoporosis and other distribution of fluids in your body that happen at anything less than one unit of gravity. Some of the solutions to that could be used on Earth for osteoporosis, but some of them would be unique. I think we know enough about physiology, or could use our current foundations, to really seriously address those two issues and others. Our microbiomes could be rethought. So could some of the neurobehavioral components of living in close quarters.
Naam: Changing how our brain works to be better suited to surviving on a small spaceship for a very long period, or in a small-quarters Mars colony for a very long period of time—what would those changes look like?
“The lesson of evolution is that there’s constantly room for improvement.”
Church: Oh, I don’t think that the final list of genes is in. We have a short list of genes for longevity and aging reversal, from many animal studies. Some of those will be relevant. We have a short list of cognitive- and anxiety-related genes that have been shown in animal studies as well. One or two or three genes will have a large effect, even though you know there are thousands of genes involved in natural populations. With synthetic biology, you’re not limited to the exact alleles or allele frequencies in the natural population.
For osteoporosis, the pathways are understood for osteoplast formation, the cells that build up and break down bones. The genes involved in calcium, metabolism, vitamin D—these are understood. With a little trial and error, we could get it so that animals are not suffering from bone loss. In terms of the fluid distribution, that I know less about, but I’m fairly sure that could be compensated as well.
Naam: You’ve mentioned longevity. How can we affect the human life span and health span through genetics?
Church: A lot of what makes young cells young is their commitment to doing repair at a good clip.
There are 300 genes that are in Pedro de Magalhães’s GenAge database4 that are up for grabs, and we’re looking at a fair number of them in the gene therapy context, so they can be applied to older animals and older people. A huge fraction of what we’re going to die from in industrialized nations are diseases that don’t kill 20-year-olds. But probably 90% of us will die of such age-related diseases. And if you get some gene therapies that get multiple ones at once, then you’re probably on the right track for something that’s dealing with the fundamentals of aging rather than just alleviating symptoms.
4 João Pedro de Magalhães of the University of Liverpool leads a collaborative effort called Human Aging Genomic Resources, which includes GenAge, a database of genes that are known to play roles in aging.
Naam: You’re talking about heart disease, cancer, diabetes, Alzheimer’s, stroke.
Naam: And there’s a possibility of gene therapies that would reduce the risk of those.
Church: Right. Or possibly, in some cases, reverse some of the disease state as well. We know the nine pathways of aging5, and there might be some relatively small numbers of genes that are highly leveraged in those nine pathways. So you just convince the cells that they’re younger, and they need to repair. You don’t necessarily micromanage the repair, you just convince the cell to do its job.
5 European researchers, writing in Cell in 2013, said these nine mechanisms are “common denominators of aging,” especially in mammals: “genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, deregulated nutrient-sensing, mitochondrial dysfunction, cellular senescence, stem cell exhaustion, and altered intercellular communication.”
Naam: Let’s go back to the slightly more far out. You made this comment recently: “The human brain may not be at the ideal size.” Can you elaborate on that?
Church: It’s unlikely that any aspect of our bodies or ecosystems or the environment is fully optimal. The lesson of evolution is that there’s constantly room for improvement. In the evolutionary sense, whatever it was that they were trying to do was to maximize procreation. There’s a whole new set of criteria that we have today. A lot of the maximized procreation of the past assumed that food was an extremely limited resource. Hence, energy had to be conserved, so you wouldn’t waste energy on things like repair, especially after you reproduced. We can reoptimize these systems.
Now, back to the brain. If we did start doing adult augmentation, there might be a desire to have a more flexible brain size, just to put new components in. It’s not to say that brain size correlates with anything desirable, it’s just that, in the transition time, you might say, “Let’s just make it bigger.” For a while, we made computers a lot bigger. Now, smaller computers are more desirable. But during transition times, you’re just very pragmatic about it. I don’t know what the right size is but I think functionality will dictate it.
Naam: If you wanted to add functionality to your brain, what kind of upgrades would you be seeking?
Church: I don’t think eliminating sleep, though some people put that on their list. I don’t think that’s necessarily a plus. I think what I would like is more working memory. The lesson we get from computers is the more things you can keep in RAM or even faster parts of the memory, those are advantageous.
Naam: Hold more concepts at once in your mind.
Church: Right. Right now, we see everything in two dimensions, we translate it into three dimensions. We can mathematically talk about four dimensions, but what if that were something that we could more deeply feel? I think that would be more interesting. We talk about consciousness, different levels of consciousness. I know late in the day I feel semi-nonfunctional. What if my very best part of the day was just my average, and then there’s something beyond that? I think that would be something that you could imagine just extrapolating a little bit, or maybe a lot. Caring for other people: it would be nice to be able to crank that up without necessarily putting one at risk for people who don’t care about you.
The list could go on and on. I’m sure that if you get a lot of creative people around the world thinking about this problem, they will come up with cool ideas and also what’s wrong with those cool ideas, how to fix what’s wrong with those cool ideas, et cetera, et cetera. That’s why we need a culture of science fiction coupled with a culture of turning science fiction into science fact.
Naam: Speaking of which: You’ve got an idea for sending small amounts of biological machinery ahead of us to explore or help colonize far-away worlds in space.
Church: Right. It’s very difficult right now to get even reasonable-sized electronic probes going at relativistic speeds6. If we go at current rocket speeds, even with gravitational assistance, it’s going to take us millennia to get to some reasonable place outside of our solar system. So, you really want a strategy by which we could go as close to the speed of light as possible. Obviously, light goes at the speed of light. And if we had a 3D printer at the other end, in principle, we could transmit something to it with copies of ourselves. We’re a bit away from making copies of ourselves. But how would you get that 3D printer at the other end, assuming there is nobody else in the universe? Until we find out there’s somebody there, we need to put our own printer up there.
6 Einstein teaches us that relativity always applies. But here Church is talking about speeds that would be a sizable fraction of the speed of light.
What’s the smallest package that we could get going at some high fraction of the speed of light? There’s been discussion of breakthrough star shots: getting a one-gram package to do a fly-by of one of our nearest solar systems. I think a fly-by will not be that informative. What we really want is to land. And we don’t want to send a gram. We want to send a lot less than that, because if a gram hits an atmosphere at relativistic speeds, you have something that’s likely the equivalent of an atom bomb. But a nanogram is something that could, conceivably, get accelerated easily and even decelerated at the other end.
There are challenges, definitely, but a nanogram is about the size of a eukaryotic cell. We know that a eukaryotic cell, a single cell, can contain enough information to create a very complex body. For that matter, a population of bodies. It could be programmed with enough information to build a light source that could beam back, bidirectionally, establish a communication line. Then you could start moving things at the speed of light. Not matter, but information.
Naam: I love this. We send a single cell to a faraway world that would self-replicate, and build a transmitter and computational strata of some sort there as our first steps to that new world.
Church: Yeah. I think that’s the smallest package that I can think of that we can get going at the highest speed. Maybe even within our lifetime.