Shankar Balasubramanian

Our Founder on Decoding DNA

Shankar Balasubramanian, the Cambridge Epigenetix founder, is interviewed by Jim Al-Khalili for the BBC programme The Life Scientific.


Jim Al-Khalili 0:00
Hello, I’m Jim Al-Khalili, and this is the life scientific. The deal is I get to talk to some of the amazing men and women who are trying to understand our world and to make it a better place, and you get to find out what makes them get out of bed in the morning. My guest today is responsible for a revolution. The ability to read at speed, the unique genetic code that makes each one of us who we are, is transforming healthcare, helping us to understand the genetic basis of many diseases, particularly cancers, and to develop new diagnostic tests, medicines and personalised treatments. Shankar Balasubramanian invented a way of sequencing genomes that is 10 million times faster than the technology that was used in the human genome project at the turn of the century. And what’s more, it can be done on a desktop machine. The company Shankar set up as a man to make these rapid sequencing machines sold for $600 million in 2006. And the business is now worth over $60 billion. It has more than 7,500 employees and supplies laboratories all over the world. But what I love about Shankar’s story is that he didn’t set out to make a lot of money or even to make something useful. He just wanted to understand the DNA double helix in the greatest possible detail to reveal how it worked molecule by molecule. Having said that, when he spotted an opportunity to create a technology that could transform healthcare, he jumped at the chance. So Shankar Balasubramanian, winner of the highly prestigious Breakthrough Prize and life sciences in 2022. Welcome to the life scientific.

Shankar Balasubramanian 1:52
Thank you very pleased to be here, Jim.

Jim Al-Khalili 1:55
Well, shank, as I say, what I love about your story is that your intense curiosity about the chemistry of DNA led to this fantastically useful machine, this fantastically wonderful technology.

Shankar Balasubramanian 2:08
Yes, it was, in many ways, it was an accident of blue skies research. Because research by its nature is hard to predict. And often when you you go out into into new space, in research, you start to see things or think differently about things. And that can often lead to unexpected opportunities. And this was this was such a case and

Jim Al-Khalili 2:33
that’s it was unexpected, wasn’t it? You know, you had no inkling at the time that this would lead to something that was going to revolutionise medicine in this way?

Shankar Balasubramanian 2:42
Not at all. In fact, we we wrote a research proposal that was funded by UK Research Council and nowhere in the proposal did we mentioned sequencing that was that was absolutely not what we were trying to do.

Jim Al-Khalili 2:57
Well, Shankar Balasubramanian before I find out more about this revolutionary technology you invented in sequencing genomes, want to find out a bit more about you? You arrived in the UK with your parents when you were just a few months old?

Shankar Balasubramanian 3:10
I did. I was born in a city used to be called Madras in southern India. It’s now called Chennai and my parents came over in 1967. I was just a baby at the time. And eventually, we settled in the north of England and that’s that’s where I grew up in rural Cheshire, I believe it was indeed it was. There weren’t many people living around there. Had a few good friends. There weren’t many children spent most of my childhood really sort of climbing trees and kicking a football and really, we were sort of blissfully oblivious to what was going on in urban places and around the world and felt like we had all the time in the world.

Jim Al-Khalili 3:53
And what about the school that you went to

Shankar Balasubramanian 3:56
the primary school I went to was a was a 500 year old village school does re school with not many children. I think there were about 50 children there altogether. And fond memories of an amazing headmaster, Brian leach was his name. And to this day, he’s he’s a great personal inspiration for me, because he made the whole experience of learning enjoyable, and in many ways, kindled my curiosity to ask questions about things very early on, in my education, and did your parents push you to they make you work hard, but you know, the classical set of Indian parents are fairly pushy, and I think my parents were actually unusual in this regard. I was very much raised in an environment where they allowed me to find myself and find my own pathway and drive myself. So they fostered my curiosity in a very subtle way. But never did they place expectations on me to get these grades or achieve anything else, they really just allowed me to relax and find my own pathway.

Jim Al-Khalili 5:13
Well, let’s bring you up to date in terms of education. If you did well at school, you ended up going to Cambridge to study natural sciences. You did a PhD at Cambridge in chemistry, then you’ve spent some time in the US Pennsylvania State University before coming back to the UK and back to Cambridge. What made you decide to focus on studying DNA?

Shankar Balasubramanian 5:34
Partly, you cannot miss DNA and Cambridge, the history of the place and the legendary figures who walked the streets.

Jim Al-Khalili 5:43
Yes. It’s not just Crick and Watson. But in fact, past guests on this programme,

Shankar Balasubramanian 5:48
very many, very many. So of course, you can’t avoid thinking about DNA if you’re in Cambridge. It’s also such an amazing molecule. I’m a chemist, I study molecules, why would I want to work on any other type of molecule?

Jim Al-Khalili 6:01
It’s the molecule responsible for life itself, after

Shankar Balasubramanian 6:03
all, absolutely. Okay, so

Jim Al-Khalili 6:05
you decided to work on DNA? And you teamed up with David Kinnaman, who was another relatively young, new young academic in the chemistry department at Cambridge? How did you both get together?

Shankar Balasubramanian 6:15
Well, this actually started because I needed a laser. And I’m an organic chemist, so I didn’t have lasers lying around. And someone told me there’s this chap klenner Minh, who has lots of lasers in his lab, why don’t you go talk to him? So I did, and he helped me. And then we continued the conversation. And this was much more exploratory. And in many ways, we had lots of time on our hands and space and capacity to try out new ideas. So we had an idea to watch a DNA polymerase.

Jim Al-Khalili 6:51
This is a We should say it’s an enzyme that helps DNA make copies of itself.

Shankar Balasubramanian 6:55
Yes. So DNA polymerase is it’s a protein, and a DNA polymerase catalyses arguably one of if not the most important chemical reaction in life. That is it stitches together the building blocks that make up DNA. And David had these new techniques for observing single molecules, which was quite new at the time. So we came up with a an idea and wrote a proposal to watch single molecules of this machine making DNA that was attached to a surface

Jim Al-Khalili 7:31
so that you were using this new technique that you were developing to study this protein the polymerase, at what point did you think you could also use it to sequence the genome to read DNA?

Shankar Balasubramanian 7:44
Well, we were actually immobilising a molecule of DNA to a surface and we were using fluorophore. So fluorophores are molecules that light up in different colours, if you if you radiate them with a laser, and we were using fluorophores, to label building blocks, and to watch them being incorporated. And we realise that if you were to colour code, all four of the bases, so DNA has four building blocks that we abbreviate to G, C, T, and K. And if you colour code, each of them differently, so the colour tells you the identity.

Jim Al-Khalili 8:25
And these are these building blocks are molecules in their own right, they are molecule atoms of hydrogen and oxygen and nitrogen cow slowly,

Shankar Balasubramanian 8:31
they are molecules that make up DNA. And what the machine does is stitch them together in a particular order, right. And that order is the sequence of the DNA. And if you were to colour code, all four of them and watch them go in, actually, you’d be inadvertently reading the sequence of the template. So putting that together with parallelization, we did some calculations that suggested we could sequence gazillions of bases

Jim Al-Khalili 9:02
was was it an aha moment? You know, or a gradual realisation that this this might work?

Shankar Balasubramanian 9:10
I think it was, it was step wise, I think in science, eureka moments are actually usually spread out over an extended period of time and a series of steps. But But what really brought together the concept that this this could be transformative is when we did calculations to suggest how much DNA could we sequence. Now a human genome has about 3 billion building blocks of information arranged in a sequence per copy. That’s a lot.

Jim Al-Khalili 9:43
This is just 3 billion of these ATCG.

Shankar Balasubramanian 9:46
Exactly It is 3 billion letters organised in a particular sequence on 23 chromosomes per copy.

Jim Al-Khalili 9:53
If you were to print out the whole genome with all these letters, it fills like over 100 Hefty volumes books. It’s a long string of letters,

Shankar Balasubramanian 10:03
really as a long string of letters. And the Human Genome Project was taking place at this time near Cambridge, at the Sanger Institute. And we were aware that this was a big scale project, and are our calculations suggested it may be possible to sequence a human genome in a matter of days, rather than, rather than, well, the Human Genome Project, of course, it was the very first copy of the genome. So it was harder than resequencing it. But that project took about a decade. And it involved many genome centres around the world to do it.

Jim Al-Khalili 10:43
So when you realise that, what did you do next?

Shankar Balasubramanian 10:46
So we some time shortly after that, we paid a visit to the Wellcome Trust Sanger Institute to learn more about the Human Genome Project. And we met three of the UK leaders of that project in a coffee room in the building. And so we shared with them our thoughts and ideas and concepts on how this might be possible to do in a day on a single system. And just just really to gauge almost to say, would this be useful? And of course, they were hugely enthusiastic about it.

Jim Al-Khalili 11:23
I mean, I’m intrigued by that meeting at the Sanger Institute, you and David, you’re still in your 20s. There, you’re just starting off in your in your careers. Clearly you excited this is going to work. Did you feel that you were you’re going to be able to convince people straightaway, they’re going to see just how wonderful it was?

Shankar Balasubramanian 11:39
No, that wasn’t a given. It wasn’t a given actually, I think partly because we were chemists, not geneticists, we were early stage untenured. no track record.

Jim Al-Khalili 11:51
Were you nervous, you know that you are telling these experts? How they could do things so much better as a sequence the genome so much more quickly? No, we

Shankar Balasubramanian 11:59
weren’t nervous. I think we we have clarity on what we were thinking about. We just didn’t understand fully the implications of this, if it were to work. And it was wonderful to get their endorsement and support. You know, they were really excited by what we were proposing and strongly, strongly urged us to go ahead and make it happen.

Jim Al-Khalili 12:26
And say, reducing the time for sequencing a full human genome from a decade, just a few days, or how much more quickly can we do it today?

Shankar Balasubramanian 12:35
Well, today, the high end instruments that use this approach. Were running flat out, they effectively sequenced the equivalent of a genome an hour. Wow. So they will sequence many genomes in 24 hours.

Jim Al-Khalili 12:54
And for a fraction of the cost of, for example, what the Human Genome Project. The

Shankar Balasubramanian 12:58
Human Genome Project, of course, was the very first genome so they didn’t have a template to go on. But that cost on the order of billions of dollars, and today, a human genome is is below $1,000. And there is a suggestion that it will come down in cost even more. Well,

Jim Al-Khalili 13:19
Shankar Balasubramanian? You set up a company Celexa, why did you think that was necessary?

Shankar Balasubramanian 13:25
Well, you know, these experiments showed that it wasn’t impossible to build a working technology that others could use. But we hadn’t done that. In order to do that we realised you would need to really bring together engineers, chemists, molecular biologists, computational people, software people, and integrate all these skills together to solve all of these parts in a way that would work as a system. So, again, we were two young untenured academics with a relatively small research effort. We actually couldn’t see too many ways of mobilising the resource needed to make that happen at the time. And I had had a chance encounter in one of the colleges actually, with an investor group. And we had a nice dinner and they had said, if you have any ideas that you think might be good for a tech company, give us a call. So David, and I call them and we came to London and described the idea literally on five pieces of paper,

Jim Al-Khalili 14:37
even you’re still in your late 20s At the time,

Shankar Balasubramanian 14:40
absolutely. And that they could see it was a high risk project to build the system, but they could also see that if it worked, it could be transformative. So they gave us the funds to start this. So really we started a company which we called Celexa as a vehicle to mobilise is the capital the resource and build the team that would be needed to actually make this happen?

Jim Al-Khalili 15:07
And here we are two decades or more later. And your technique for sequencing the genome is now revolutionising medicine, isn’t it?

Shankar Balasubramanian 15:15
I think the technology has has really had a bigger impact than I think I certainly imagined it’s taken a bit longer. But the impact is bigger,

Jim Al-Khalili 15:25
must be gratifying. It I feel

Shankar Balasubramanian 15:29
very fortunate to have been involved with a concept idea that has inadvertently done something quite useful.

Jim Al-Khalili 15:37
Maybe you can say something about what the tangible benefits are to health from using this technique?

Shankar Balasubramanian 15:43
Well, there are there is a concept of personalised medicine or personalised health care. And large projects such as genomics England NHS project in the UK, I think have been fabulous for both doing the research and building knowledge but also helping patients.

Jim Al-Khalili 16:02
So genomics England, this is a project that sequencing the genomes of NHS cancer patients, and this is helping us all understand what’s happening to a person’s DNA when they develop a particular kind of cancer.

Shankar Balasubramanian 16:15
Indeed, and cancer is a disease caused by changes to the DNA. And by picking up those changes through sequencing, one can learn in great detail exactly what the underlying traits are of the cancer. Another key area is actually rare. Diseases. Rare Diseases are often so cool, because they’re hard to classify. And they’re relatively unique. But as a collective, it’s one in 17 people.

Jim Al-Khalili 16:46
Right? Yeah. You’ve talked about what we can do with healthcare in terms of sequencing the human genome. But of course, what’s been in the news recently with the pandemic, is that we’ve relied on rapid genome sequencing to identify a virus, the COVID cough two virus, which is and its variants and be able to predict what will happen next.

Shankar Balasubramanian 17:07
Think, broadly speaking, infectious diseases is the third area of medicine where sequencing is having a major impact.

Jim Al-Khalili 17:16
There seems to be just so many uses for these rapid sequencing machines that you’ve helped develop. And not surprisingly, your company Celexa, which later became Illumina is highly successful, worth many billions of US dollars, so a lot of money and suggests I guess, you must be quite a wealthy man.

Shankar Balasubramanian 17:37
What I’ll say is, when when David and I started the venture, we certainly didn’t optimise for our own wealth. If we were to start a venture today, we might we’ve learned a bit more. But I still go to work on a rusty old bicycle, actually. So

Jim Al-Khalili 17:58
you must have put some money aside for your retirement, presumably, we don’t have a mortgage. And I hope you’ve retained some, some interest some shares in the company,

Shankar Balasubramanian 18:07
I’ve, I believe in the long term vision, and that also applies to my whole life.

Jim Al-Khalili 18:15
So it’s you’re not the only UK scientist I know, who’d rather not talk about money, but turning ideas into useful technology, having a successful company generating wealth, creating jobs. These are all good things, surely something to be to be proud of.

Shankar Balasubramanian 18:30
I think what I’ve seen over the last 20 years is building an ecosystem in the UK and technology areas has sort of long lasting ramifications. So the UK is actually one of the world’s leaders in genomic technologies, and the application of genomics. And there are there’s more than one successful genome technology company in the UK.

Jim Al-Khalili 18:56
So far, Shankar Balasubramanian, we’ve talked about the genetic alphabet, those four letters A, C, T and G, that make up this code book of life. And we’ve known about this ciphered for decades now. But it turns out, there are more than these four letters in the genetic alphabet. A fifth letter was discovered some years ago, and more recently, a sixth seventh, and an eighth letter had been identified. What got you into the study of these new letters, these so called epigenetic letters,

Shankar Balasubramanian 19:25
it was a chance encounter, I would say, one of my senior colleagues in Cambridge, Tony green, who’s a haematologist called me in my office one morning, and said, Have you looked at the latest edition of science? And I said, No, I haven’t. But I’ve got it right here on my desk, and he told me to turn to a certain page number. And there was an article announcing that a new letter had been discovered and shown to exist in human DNA, a variant of the letter C and you You know, he said this is in DNA. So it must be important, but we don’t know how to sequence it. Do you think you could come up with something? And I had a look at it in real time and thought about the chemistry of this epigenetic letter. And I said, I can’t tell you the answer now. But it looks like a solvable problem.

Jim Al-Khalili 20:21
And as before, for you the challenge here was to find a chemical reaction that was unique to this letter that would identify this letter.

Shankar Balasubramanian 20:31
Absolutely. And that’s the sort of problem that chemists love. Something that’s important. That requires new chemistry. And and if you solve it, you will sort of provide a means to liberate the generation of better knowledge and understanding about that.

Jim Al-Khalili 20:48
And of course, you did succeed quite quickly. In fact, within two years, you’d worked out how to identify the SIG letter. And you set up another company called epigenetics to help develop the technology to see these new epigenetic letters. What have we learned as a result of this, this new insight into the into the structure of DNA?

Shankar Balasubramanian 21:10
Well, firstly, yes, the work took two years, it was just myself and one PhD student actually worked on that problem. And we filed a patent and started a company called Cambridge epigenetics. And in a sense, this was the epigenetic equivalent of Celexa it was a chemistry for sequencing, not just the genetic basis, but the epigenetic basis. So you

Jim Al-Khalili 21:35
so it’s really is more information that you’re extracting from DNA. It’s not just the instruction manual to encode all the different proteins, but it’s the additional instructions to tell those proteins when and how to do their job.

Shankar Balasubramanian 21:51
Indeed, and the the epigenetic information is dynamic, that is it can be changed. And it can change in a way that’s adaptive, it can change in a way that’s related to the identity of the cell. In fact, there are concepts that suggested changes as cells and organisms age, as well. And it can shift during the onset of certain diseases. So there is there is additional information to mind right from the epigenetics that reports on the biological system,

Jim Al-Khalili 22:29
what would you say, we’ve learned from the study of this new epigenetic information using the technology you’ve developed? What what we’ve

Shankar Balasubramanian 22:37
learned from from epigenetics is, epigenetic changes can sometimes tell you things that are different from what the genetics tells you. And in diseases such as cancer, actually, there’s now some exciting data coming out from organisations that use blood based testing of DNA. By measuring the epigenetic patterns in DNA floating around in blood, you can detect cancer.

Jim Al-Khalili 23:10
So we did detecting cancer at an earlier stage. I mean, this is a this is a revolutionary new way of detecting cancer.

Shankar Balasubramanian 23:16
Well, the ambition, not just in cancer, but I would say in all diseases is early detection. And earlier detection could potentially lead to prevention or more successful cures. So this is the aspiration. And that there is some suggestion from the biology that detecting epigenetic changes might be a way to catch the onset of disease earlier. Very,

Jim Al-Khalili 23:43
very exciting. To recap for a moment that you have these original genetic letters, ATC and G, the Fab Four. That’s one code one way in which information is stored in DNA. Since we’ve cracked that code, more letters have been discovered fifth, sixth, seventh, and so on, you found a way to crack that code to identify those, those chemicals. Now, as well as these epigenetic letters, you think there could be another way in which information is stored in in our DNA? What’s that?

Shankar Balasubramanian 24:14
Well, I’ll start by saying, DNA has been around for a very long time. And features that occur quite commonly in DNA across organisms need to be looked at quite seriously, I would say because it’s likely that they’re doing something important and interesting. So we’ve dealt with genetics, which is if you like the the primary structure or sequence, EPI, genetics is a another dimension of chemical molecular information. We’ve also in my group spent over two decades now looking at a folded structure of DNA. That is an alternative to the classic double helix that everybody is familiar with. So In the double helix, you’ve got two strands of DNA twisted together. It’s been known for some time that odd sequences of DNA that have lots of the G letter in them can, in a test tube, spontaneously fold into a four stranded structure. That’s right. It’s rather like a well ordered knot.

Jim Al-Khalili 25:21
But you don’t yet know precisely what the role of these knots in the strands of DNA is.

Shankar Balasubramanian 25:27
What I would say is that we have some good ideas based on data. We haven’t nailed it yet. But it appears as though they are involved in controlling whether genes are switched on, or switched off. So in many ways, they act rather like the epigenetic letters, and may be helping orchestrate what’s happening to all the genes in the genome.

Jim Al-Khalili 25:51
When we think back into the textbook example of Crick and Watson’s double helix structure of DNA and just how revolutionary that was in in science going back to the early 50s. Do you imagine these these new structures, these knots in the strands of DNA would equally revolutionise our understanding of this molecule of life?

Shankar Balasubramanian 26:10
Well, I think it would take a very bold person to say something equals what the work of Watson Crick and Rosalind Franklin has taught us, because that has taught us about the fundamental mechanism by which genetic information is stored and transmitted. What I would say though, is, DNA is a remarkable information molecule. And it has many layers as to how it stores and transmits information in biological systems. What we and others in the field are doing is discovering some of these other layers and mechanisms of information that DNA has and figuring out ways to fully decode everything that DNA is telling us. And for

Jim Al-Khalili 26:59
you, as a young man, thinking about what you wanted to focus on in your research, and deciding to study DNA, it hasn’t let you down has it?

Shankar Balasubramanian 27:08
DNA has never failed to keep me excited and curious. It’s on the face of it relatively simple molecule. It’s just four building blocks arranged in a particular order. But But actually, rather like an onion as you as you as you learn one thing you see other things, and I think there are more surprises for those of us who are looking closely at DNA.

Jim Al-Khalili 27:33
And for you looking closely at it, you haven’t let us down in terms of turning it into something that benefits humanity.

Shankar Balasubramanian 27:39
Well, DNA and of course, it’s closely related RNA dictate everything in every living system. So how could you not learn more about life by studying it?

Jim Al-Khalili 27:51
Basically Watch this space, Shankar Balasubramanian, thank you very much for sharing your life scientific. Thank you. And thank you for listening. I’m Jim al Khalili, the producer is Anna Buckley. And if you’ve enjoyed listening and would like to hear more of my interviews with leading scientists, why not subscribe to the life scientific? It’s available on BBC sounds, and subscribing will give you easy access to interviews with more than 200 men and women who are changing our world.

Cambridge Epigenetix is now biomodal