Pushing the limits of genomics

Published Mar 3, 2024, 8:33 AM

Genomics is fundamentally changing the way we think about health and disease and driving forward personalised treatment options that would have been pure science fiction just a few years ago. Still, some parts of our genome are more ‘mysterious’ than others. In this episode, we meet Garvan researcher Dr Ira Deveson who is spearheading cutting-edge DNA sequencing methods to future-proof how genomics will integrate into healthcare.

Genomics is fundamentally changing the way we think about health and disease and driving forward personalised treatment options that would have been pure science fiction just a few years ago. Still, some parts of our genome are more mysterious than others. Garvan researcher, Dr Ira Deveson is spearheading cutting-edge DNA sequencing methods to future-proof how genomics will integrate into healthcare.

You're listening to Medical Minds, the podcast that takes you inside the labs at the Garvan Institute of Medical Research. I'm your host, Dr Viviane Richter and with me here is Dr Ira Deveson, Head of the Genomic Technologies Lab at Garvan. Welcome Ira!

A. Hi, Viv, thanks for having me. Really looking forward to our conversation.

Ira, what excited you about genomics enough to make it your career?

Yeah, I guess I've always been interested in the life sciences and the natural world and I guess just sort of understanding how things work in the world around you, which, which really is the goal of science. And I think I was lucky enough to have a couple of really excellent high school science teachers towards the latter stages of my high school who kind of inspired me to take up biology and chemistry at university.

And I think somewhere along the line there, during those years, I realised, if you want to understand how the world works, how living things work, how they came to be and um how they're changing, you really need to understand the genome because it is the core program of life that dictates everything. And I guess once I'd realised that there was never really gonna be another path for me, I was always fascinated by it from that point onwards.

Tell us about the first time you stepped into a research lab.

The first time I stepped into a research lab would have been my honours year at ANU.

So during honours, I studied plant science, doing a plant genetics project. So I was embedded in a lab where we were growing little model organisms called arabidopsis and fiddling around with their genomes to test things about how their genes operate and switch each other on and off. My project was actually a bit of a left field one where we decided we wanted to clone a human gene into this plant.

We were testing the idea that this gene is a really ancient gene, its function has been around for so long. That if you take it from a human cell and put it into a plant, it could still perform its function. We did that and it kind of worked, it kind of could still perform its job. But the plant was a bit messed up. It sort of had weird curly leaves and was a bit shrunken and imperfect.

But it was a really fun experience about what you can do in science when you come up with a hypothesis, even if it's a bit left field and just have a crack at it. Yeah.

What did that tell you about genomics?

I mean, it taught us a lot about this gene. It taught us that the gene

is so deeply conserved or so ancient that it could still function in a plant cell. But obviously, there had been in the intervening hundreds of millions of years, it had changed sufficiently that its job wasn't perfectly executed in a plant cell. It wasn't interacting with all of the things that it would normally interact with in a human cell. So, yeah, we learned something about the evolution of that gene which to me was eye-opening and a pretty cool experience as a undergraduate.

So you went on to do a PhD. Can you tell us about what you did?

Yeah. So my PhD was in genomics in a pretty broad sense and it was mostly sort of computational stuff. So I'm a bioinformatician or, or I was a bioinformatician , a person who analyses genomics, data sets to learn things about biology and, I say that because it was a very broad PhD, I worked on a lot of different projects, all of them on different subjects, but united by the fact that they involved genomics.

Actually, the title of my PhD thesis ended up being 'Three largely unrelated experiments in the era of genomics' or something like this, which my supervisors were not very impressed with. But I think the main thrust of what I was doing turned out to be a a biotechnology project where we developed a new product which was called sequins. These are sort of synthetic genes. So totally artificial molecules that we created

that you could actually use as internal controls in your genomics experiments. So the basic idea is you would get these synthetic genes, you'd add them to a DNA sample that you were just about to analyse like maybe a patient DNA sample. And then you'd do your genomic experiment as you normally would, but you'd have these synthetic molecules in there that act as kind of like an internal truth set or like a a known and

within your experiment and having that in there, lets you sort of measure how accurate your experiment is because you've got this truth set, you can see if you're making any mistakes and that gives you ability to sort of do quality control, measure your performance over time. And ideally to optimise your performance, you actually improve the accuracy of your experiment.

So rather than investigating a specific biological problem, you're more focused on developing the technology that sits behind that.

Yeah, I think that's what sort of came out of my PhD was this philosophy that you don't necessarily have to be a expert in one particular disease or one particular biological system. But you can spend your time building tools and, and methods and get these sort of capabilities particularly in the data analysis, which is a huge part of genomic science.

And once you've got this expertise or once you've built those tools, you can really apply them in many different areas. And the applications sort of follow naturally, I guess, usually via collaboration with people who are experts in different disease areas or different areas of biology.

What sort of sequencing do you specialise in?

So my team's called the Genomic Technologies Lab. And we work on lots of different genomic technologies, but something that we're really interested in at the moment and over the last couple of years is called long-read sequencing. So this is a new class of technologies that are really hitting their stride over the last couple of years and they're kind of revolutionary because I guess over the last 10 or 20 years in genomics, we've been pretty reliant on a technique called next-generation sequencing or shotgun sequencing.

And this is a process where you basically smash the genome up into tiny little fragments that are about 100 base pairs long or 100 letters of the DNA code long. The genome is 3 billion letters and we're smashing it up into these tiny 100 base pair pieces. And then you have to stitch all those together like a, like a jigsaw puzzle to sort of understand what the genome looked like in a given person or a given organism.

But now with these new technologies that are coming along, long-read sequencing - it's called long read because we don't have to smash the genome up into these tiny pieces. We can actually read much longer pieces of the DNA code. So thousands or tens of thousands or even up to millions of letters long in a single continuous piece.

So that basically just means that when you're stitching the genome back together for your patient or for your organism, you're doing a much less complicated jigsaw puzzle. You don't have all these tiny little fragments, you have these long continuous molecules. And so that makes the problem much simpler, much less chance of making mistakes. And it lets you access parts of the genome that are too complicated to stitch together with those small jigsaw pieces.

You're saying this is giving us the most accurate view of our genome.

Yeah, this is giving us the best resolution, I would say when it comes to reading a genome and

I guess you need to think about the fact that the genome is not a single homogenous object. It has lots of areas that have specific contexts that make it difficult to read. And the thing that's usually most difficult is a repetitive sequence. And you can imagine if you're doing a jigsaw

puzzle, if there's a section of the puzzle that has lots of the same feature being repeated over and over again without any kind of unique features, that's really hard for you to interpret and stitch together correctly. And that's the same in the genome. If we have large sections of the genome where you just have the same

DNA code repeating itself again and again, you're very likely to make mistakes when you're trying to piece that together and read it out. So there's maybe 10 or 20% of the genome that's like this. It's too repetitive for us to have properly analysed in the last 10-15 years of genomics. But now with these new technologies, we can finally read these sections, these sort of mysterious dark regions of the genome.

And why do we need to know about these regions?

Partly because it's just really interesting, we don't know what's in there. And that's what drives us as scientists, is driving our knowledge for just finding out what these regions look like. But we also know that there's lots of important genes within these regions and there's lots of important

genes that have roles in disease where we already know that they have roles in disease. But we haven't been able to properly analyse them because the technologies have not been there or we can analyse them maybe to diagnose a genetic disease, but it's really slow and laborious and challenging and error-prone . So with these new technologies, we can eliminate a lot of those difficulties, but we can also just shed new light on things that we haven't understood before.

An example that we've been working on a lot in the last couple of years, a specific group of disorders called repeat expansion disorders. These are typically neurodegenerative diseases affecting the nervous system, but they involve lots of different genes. There's at least 50 genes involved and probably many more out there that we haven't yet discovered.

But they contain these repetitive sequences that are kind of like repeating syllables within a word, everybody has these genes. But in certain people, those repetitive sections of the gene actually get blown out in size and they become massively expanded. So you can maybe think about a word like Woolloomooloo, the suburb in Sydney, everyone has a few repetitive syllables in that word. But maybe in your patient when you read it out, there's is Woolloomooloomooloomooloo , right? And so it's become massively expanded. And

that's the nature of these repetitive sequences that they're quite unstable, but that actually leads to disease in some cases. And so some very well known genetic diseases like Huntington's disease is actually caused by these repetitive expansions in a particular gene.

So those have been difficult to research and difficult to diagnose until recently. But we're actually now applying these new long-read sequencing technologies that we have to read these out very accurately so that we can tell the difference in the length of these genes between different individuals. And that we hope is gonna lead to much better, much more streamlined, much more accurate patient diagnoses

being able to diagnose these diseases much earlier on would lead to a transformation in patient outcomes.

So unfortunately for a lot of these neurodegenerative diseases that we're looking at, there's no current cure, but there are cures on the horizon. There are very promising clinical trials getting underway for things like gene therapies now coming out for different inherited diseases. It's only a matter of time before there will

be ways to treat these disorders that are very debilitating, but we really need to be able to diagnose them and work out exactly what a particular patient's specific mutation is or variant is so that they can be accurately targeted within these trials and when these therapies come online.

Ira, tell us what else you're using this long-read sequencing for.

Yeah. So we've spent a lot of time and energy building up our capabilities in this area, long-read sequencing. We have a lot of different instruments and we've really become the leader in Australia in this area. And it means we actually work on all sorts of different genomics projects. There's a lot of demand for these instruments and for these capabilities. But we started offering a sequencing service and we collaborate with many different researchers around

Australia. So we work on everything from clinical genomics projects like what I've just described where we're trying to improve patient diagnoses. We work on cancer research which is obviously a disease that's driven by genomic events. We work on immunology because there's a lot of great immunology labs at the Garvan, who we collaborate with, who you know, require genomics capabilities for their work.

But we also do some more left field things like working on non-human biology. So we collaborate with a lot of researchers in the plant and animal world who want to sequence a genome for their particular species that they're most interested in. Maybe for evolution studies or ecology

studies, we help them build the genomics resources that they need to make their work happen. That's a really interesting and fun side project that we do. But it often teaches us a lot about how the human genome works as well to understand non-human organisms. In addition to that, we work on different microbes including viruses, which have obviously been very front and centre over the last couple of years. And obviously, you know, all of these things involve DNA, all of them have a genome.

And basically in modern science, you need to understand the genome to understand whatever organism you're working on. We've found that there's a huge appetite for these technologies. We obviously put in the hard work to build these capabilities up. But then once we've started offering them to the wider research community in Australia, there's no limit to the people who are showing up with cool projects to work on and really taking advantage of the capabilities that we have at the Garvan Institute.

Speaking of viruses, in 2020 you were approached about a project that had quite a significant impact. Can you tell us about that?

I started my new role as a lab leader at the Garvan Institute

in I think it was February 2020 as the COVID pandemic was just getting underway. And so within about a month of me taking over that role, the world had descended into chaos and we were in lockdowns. And scientifically, it was quite interesting for us because we had these DNA sequencing technologies that we were working with, that we had really unique expertise in Australia. And we realised pretty quickly that these could be applied for sequencing the gene

of the virus, which is a really useful thing to be able to do because it allows you to tell what the strain is. And as you know, we all became super familiar with Delta and Omicron and all these sorts of things. But you know, that's ultimately a genomic classification. You need to sequence the virus's genome to know what strain it is. It also informs contact tracing. So the genome of the virus acts kind of like a fingerprint and you can work out who passed it on to who. So you can connect these clusters or these outbreaks together,

like the Crossroads Hotel and the Thai restaurant in Kings Cross. To know that they had a common source, you have to be able to sequence the genome of the virus which provides that fingerprint that I mentioned. So we knew that we had the capabilities to be able to do this. We were generally using these technologies to sequence human genomes, which are much bigger and more complicated sequencing. The virus is actually not that

hard. And so we got in touch with some virology labs and also ended up getting in touch with NSW Health Pathology very early on in the outbreak in Australia. And we said, you know, we've got these technologies and we are available to help if you need it. And so that kicked off a fairly long and intense and exciting collaboration with NSW Health

to look at a work flow , to perform this very rapid DNA sequencing on the virus that we would then feed back information to them to inform their contact tracing and, and strain surveillance.

We are four years on from that time, tell us about what you're focusing on now.

So having built up all of these capabilities in these long-read sequencing technologies that I've mentioned, we've now got the ability to sequence human genomes at scale quite cheaply. And when I say sequence human genomes, because we're using this unique technology, we're fully sequencing them. So we're getting right into these dark mysterious corners of the genome that I mentioned, these repetitive genes that haven't been able to be to be read out. And these are very interesting because they're the most dynamic parts of the genome, they're the

most diverse parts of the genome. So if you compare the genome of any two human individuals, the places where they're going to be most different are in these dark repetitive regions. And so a disproportionate source of genetic variation or genetic diversity are in these genes. And so what we now want to do is actually scale this up, spread it out

to look at diverse human populations around Australia and potentially around the world to study these dynamic regions and start to build up a a much clearer picture of how they look across populations. Like what are the shared features between individuals and between different community groups? And what are the features in these regions that are unique? And ultimately, the goal is to build reference data and the resources that we're gonna need to

understand these in a medical context. So that we can potentially diagnose genetic diseases that involve these genes to do that you need a really clear picture of what that normal genetic variation actually looks like because that's just your normal healthy background level.

What have you found out through that work so far?

So it's early days, but we've just been involved in a very exciting pilot study with the National Center for Indigenous Genomics in Canberra. It's a long-term study that we're going to be continuing to work on. But the goal of this pilot phase was just to look at the genome sequences of individuals in four indigenous communities in Northern Australia. So three in the northern territory and one in Queensland

and to start to characterise genetic variation in these groups, but specifically focused on these complex repetitive areas of the genome and and what we call structural variants. So these are large pieces of DNA that are deleted or inserted or rearranged between different individuals' genome sequences. So big changes that are really often quite relevant and have a big effect on people's genetic code. This is kind of the first time that

indigenous Australian genomes have been studied to this level. And so we obviously characterise a whole lot of unique genetic diversity, a lot of new gene variants that have never been seen before and probably don't occur outside of Australia. So we discovered a lot of variation that appears to be unique to indigenous Australians much more than you would find in a European person who comes from a population that's been very heavily studied in the past.

In addition to showing that these indigenous communities are genomically very different to people from other parts of the world. We also discovered somewhat surprisingly that these communities are very, very genetically different from one another. So people from two different communities in the Northern Territory separated by a couple of 100 kilometres, shared some genetic variation, but had many many unique genetic variants that were not found in another community that were

specific to that community. They almost look like people from totally separate countries in different corners of the globe, even though they're just separated by a few 100 kilometers in the northern territory. So that speaks to how long these people have been living here in this country in largely isolated pockets of the country and, and how unique their genetic variation is.

What does this work mean for indigenous Australians?

This work tells us a lot about the genetic diversity of indigenous Australians. They're really rich and unique genetic features.

But what's really more important about this is the first step towards building the resources. We need to be able to diagnose and research genetic disease in a manner that's fit for purpose for these communities. So as I said, we discovered that each community is very different from one another and that all of them are genetically very different from Europeans or East Asians. And so if we're trying to diagnose any

indigenous person with a potential genetic disease, at the moment, we'd be comparing their data to a reference data sets that were built with Europeans. And it's not the right comparison. So you're gonna find lots of features in that indigenous patients genome that look different or stand out from the crowd if you're using a European backdrop, I guess, and

that makes it very hard to diagnose the disease because those are all gonna look like potential places where a gene might be broken or something. But it's actually just a natural feature of the fact that these communities have been living, you know, separate for hundreds of thousands of years potentially. So it's really important that we build that reference data and, and, and build that map of what normal healthy genetic variation looks like in the appropriate communities in, in, in Australian indigenous population. And even in the specific communities that particular patient might come from

genomics is the future of healthcare. And it's gonna become increasingly intertwined and embedded with, with everything that we do in medicine that might be five years down the track, 10 years down the track, 20 years down the track. But if we don't build the resources now for these communities, they're gonna be left behind. And that's just gonna be another area in which the health disparities that already exist in this country become wider.

And I understand you have a special family connection with this project.

Yeah. So it's a bit of a crazy coincidence but these communities that we're working with um and this, this center called the, the National Center for Indigenous genomics was actually started around 10 years ago. Um At the anu,

what prompted this was that a large collection of Indigenous Australian blood samples were found in a a minus 80 freezer at the John Curtin School of Medical Research that had been basically collected in sort of the 19 I think 19 sixties. So,

uh and then how is it the anu indefinitely just sort of put on ice uh because the the person looking after them and, and running the project was no longer around these samples were discovered. And the National Center for Indigenous genomics was set up to basically repatriate them, return them to the the communities from where they were collected,

but also to prompt dialogue with those communities so that we could then start working with them genomic and come up with projects and engage their leadership for future genomics research. Some of the samples actually ended up being analyzed some of these historical samples. You can still get DNA out of them and you can uh you know, with the consent of the of the communities involved, you can still do genomics research on them.

The real crazy coincidence here at these samples were actually collected by my grandfather in the, you know, who, who was a, a genetics, genetics researcher back many decades ago. He died when I was quite young. So I had no idea who was working on any of this stuff. My mum was sort of vaguely aware and she was obviously contacted when they were starting this center

and utilizing his old lab, notebooks and journals and specimens and things. But then just much later on, I was just at a conference and my colleague from Anu was giving a talk about this new center and popping up photos of like historical photos that my grandfather

taken in the field of him working with, with these communities. And I was like, oh that's, that's my, that's my grandfather and he was like, oh, bloody hell. So, uh, we, we better, we better get you involved in the, in the research. And so that was, that was kind of how we struck up the conversation. And, yeah, just a, a bit of a crazy coincidence.

I a before we let you get back to your genomic sequencing.

It's time for the fast five. What do you do in your downtime? Uh,

at the moment I'm doing a lot of long distance running. Most

challenging thing you've ever done. When

I finished university, I rode my bicycle across Europe for about 4000 kilometers over a couple of months, which was extremely challenging. I wouldn't say it was something I had to do. I did it voluntarily and it was a lot of fun. Uh, but yeah, that was pretty difficult. Any

secret skills.

Uh, I'm really good at trivia questions about landlocked countries. So, if you tell me any country, I can tell you whether it's landlocked, possibly Kenya, not landlocked. Yeah, 100% sure about that. There's two countries that are doubly landlocked. So they're landlocked by other landlocked countries. Uzbekistan and Liechtenstein. You need to know that one. It, it comes up on the, on the quiz quite a bit.

Any pets? Uh, yeah, I got a dog called Tony. Uh, short for rigger. Tony. He's a, uh Maltese terrier cross with a toy poodle. So a moodle. He's a good little Boy,

what's the current book you're reading?

I'm reading a book called Paradise Estate by an author called Max Easton. He's a Sydney author. The book's actually about a moldy sharehouse in Sydney's Inner West, which is a topic. I'm very familiar with

Doctor Ira Davison. Thank you so much for joining us on medical minds.

Thanks for having me viv It's um been really great to speak with you if

you'd like to know more about Ira's research or donate to the work we do at Garvan head to garvan.org dot au. And if you've enjoyed this podcast, please leave a review and share with other podcast lovers. I'm Doctor Vivian Richter. Thanks for listening.

This podcast was recorded on the traditional country of the Gadigal, people of the Eora nation. We recognize their continuing connection to land waters and community. We pay our respects to Aboriginal and Torres strait islander cultures and elders past present and emerging.

In 1 playlist(s)

  1. Medical Minds

    26 clip(s)

Medical Minds

We dive deep into the minds of our amazing researchers to find out how they tick and how they are wo 
Social links
Follow podcast
Recent clips
Browse 25 clip(s)