Subscribe
Share
Share
Embed
Documenting the monumental discoveries of researchers, and the role Pawsey plays in supporting their breakthroughs.
Welcome to HPC Hearts & Minds, where we talk to the people behind some of today’s most fascinating discoveries.
High Performance Computing, or HPC, allows researchers to tackle incredibly complex questions. This series is all about finding out the real-world impact that comes from using this large-scale computing.
We want to showcase the hearts and minds behind the technology.
Pawsey sits on the Whadjuk country of the Noongar Nation and we'd like to pay our respects to their elders past, present and emerging.
It's trying to work out that now we sort of have an idea of what's present in the ticks and in the wildlife.
Can we find something in the humans that may be causing that illness?
Can a single tick bite change your life?
Welcome to HPC Hearts and Minds, where we talk to the people behind some of today's most fascinating discoveries.
High performance computing, or HPC, allows researchers to tackle incredibly complex questions.
This series is all about finding out what can be done using this kind of large-scale technology.
More importantly, it's the hearts and minds behind that tech that make it special.
In this episode, we meet Dr.
Siobhan Egan, a research fellow at Murdoch University who studies the microbes, parasites, and pathogens that move between animals and humans.
She started her career tracking wildlife through the Australian bush, and today much of her work happens at the command line, using supercomputing to analyse enormous genomic data sets.
From uncovering new species of microbes to investigating mysterious illnesses linked to tick bites, her work is helping researchers understand how diseases emerge, spread, and evolve, and how new technologies could one day identify dangerous infections in real time.
This is a conversation about curiosity, computation, and the tiny organisms that can have enormous impact on our lives.
Siobhan, thank you so much for joining me in this lovely, lovely space.
This is incredible.
So I guess for people who are unsure of your experience, could you give us a little bit of a rundown of how you first got involved in, is it parasitology or parasitology?
Parasitology.
Parasitology.
Having said that, there's
a lot of different people say it different ways.
So anyway, it's good when you go to a conference, everyone knows what you're talking about.
The ones that make people half the way and be like, parasites.
Yes, I know what you're talking about.
So a lot of what we're doing is trying to describe the different parasites and the different pathogens that are present in our wildlife that can either one, make the wildlife themselves sick, or the potential to jump to humans and make humans sick,
either now or potentially in the future.
So I think as well having that catalogue of what's present is really important for when outbreaks happen like COVID, we're able to trace back to see what the nearest sequence is that we've described before and where was that animal, where was that previously found before so that we can untangle those webs about the transmission dynamics and the evolutionary origin of that disease.
I started my degree in animal health.
So I would say I, to summarise my journey to research today, I would definitely say I came for the animals, stayed for the parasites and have continued on now with more of the data analysis side of things.
So I came back to uni as a, I guess you would call it, a mature age student.
thinking that I wanted to go down the vet pathway or something to do with animals there and ended up really liking research and did my undergraduate in animal health and wildlife conservation.
What was it about, I guess, at that point when you were going down that pathway and then you saw about this particular field that got you interested, what was the first thing that made you go, oh, this was really neat or what hooked you into it?
The animals, I guess, the capturing that wildlife and going out into the southwest bush, which I did during my undergraduate.
I volunteered on a bunch of different projects, everything from bats to birds to sea life and then sort of found my niche in the terrestrial mammal space and really liked that.
But I guess mainly I actually like the lab side of it and looking at what was under the microscope after we released these animals and took their samples.
That's what really made me stick around and find out the
answers to those questions.
Yeah.
I think it'd be funny because I guess doing all the sort of groundwork, you think, you'd imagine someone enjoying that more than the other, but it's nice to know that there's like two parts of it where you know you have to go and get the samples and get the information and then having to deal with it afterwards.
That's actually, there's just as much excitement in that part too.
I think it's the variety, definitely.
I think doing anything day to day, even if it's something that you think that you really love, anything day to day
doing it over and over again gets a bit repetitive.
So I really love the variety that research can provide and the opportunities that don't seem obvious to you at first, like doing something with HPC.
I never would have thought when I started my undergraduate journey.
What was the hardest animal to catch?
I'm curious now.
So you have to go down and you have to find, oh, we're going to find a new bat and you've never done it before.
Obviously you have help to do it.
what is that like to have to first find out where they live, and what's that process like?
So you rely a lot on the rangers and the local land carers in that area to tell you a lot about they have a lot of information that sometimes isn't published or isn't easy to find.
And then there's a number of traps you can set and it really is a luck game.
Especially in Australia, your trap success rate is really low.
Usually you're lucky if you get maybe 30 percent of your traps set.
So it's a lot of effort sometimes for what might seem like very little reward, but I think seeing an animal that no one else has seen or sometimes even heard of is really cool.
Kind of discovering new things still to this day, I guess, is quite exciting because you would feel like...
by now everything's been discovered.
It's like that's not the case, right?
Absolutely.
Especially here in Australia, there's a lot that is undescribed to Western science, I guess, in lots of ways.
And documenting that and sharing that publicly is really exciting.
So you get into that space, you go into your undergrad and you're sort of finding your way through.
Sort of yeah, what got me to that point was looking at how disease can affect the animals, but also how the diseases can jump from animals to humans.
And I started my PhD in 2018.
So that was before COVID, which seems like a long time ago.
But I remember for the first two years of my PhD, trying to describe that research to my family was really hard and they couldn't see why would you look in wildlife for human disease.
One of the upsides of COVID was it made my research really easy to communicate and people could understand quite quickly why we were doing it that way.
So that was really
nice thing to come out of COVID.
Must have been a really fascinating time scientifically for you as well, I guess, being in that space and then trying to learn what COVID, I guess, was doing and how it was being transmitted and things like that would have been really
fascinating in your field.
What later was described as the origin of COVID made sense in terms of what we find here in Australia is that things are new and they're novel and they can jump quite quickly from animals to humans in a way that may not be obvious at first.
So the origins of COVID that were heavily debated and still in some communities that are still debated really made sense for a lot of us that were doing that evolutionary biology work into pathogens and
why they jump from one host to the next.
And one thing it made my discussion very quick to write at the end of my, in about 2022, when I was writing my final discussion, it made, I guess, relatively speaking, it made it quite quick to write because there was by that point a lot already published and a lot discussed about the origins of why pathogens jump from one host to the next.
So it was cool in that respect.
But yeah, it also made it a bit of a daunting task to comb through everything and make sure that you are doing everything
you think and analysing the data in the best possible way and reanalysing again and again as well.
Did you feel like at the end of your thesis that you felt like this is definitely what I want to do or there was still some time afterwards in which you were like still wanting to
research certain different things or was this like a path where you're like, yep, this is great.
Because after a thesis, you're tired.
You've probably done it for many, many years and you're like, oh, I need to do something else.
I think like you described every person that finishes off their PhD ends up with that existential question about where they want to go.
I was, again, it was an interesting time to finish up doing a thesis in 2021 where the idea of doing an overseas postdoc quickly deteriorated with our borders being stricter and stricter.
And I was lucky enough to get a position pretty much three years after I started my PhD working on something totally different in analytical chemistry at the Phenome Centre with Professor Elaine Holmes that had just recently brought her team across from London to Murdoch and was able to join their team looking at the human gut microbiome.
So different host, but similar lab methods that I had done in the animal microbiome world.
Is it hard to jump from, I guess,
the terrestrial animals that you were thinking of working with to humans?
Was there a massive leap in having to learn a whole bunch of different types of biology or did you feel like you were already well pressed in that sort of space?
I felt like with the microbiome side of the biology I was...
I was equipped and I was okay to jump to humans.
What was another language was the chemistry to me and working with a bunch of chemists and analytical chemists and learn a whole new field.
So that definitely felt like it was a new language for at least the first couple of weeks or months.
The way in which you got into the Pawsey internship, so we're going back to 2017.
What was it about getting into that kind of internship at that time before you were, before 2018 that made you want to go into it?
Like what was it?
I should probably go for that.
So it was Pawsey at the time were coming out and doing some training for some basic command line coding and some skills to get researchers to use HPC.
And they were coming out weekly.
It was some of their trainers, including Mark Gray at the time was doing the training and it was towards the end, I think about eight weeks session.
And he sort of just dropped the idea that I should apply for an internship with them later on that year.
And I guess I held him to it.
And a few months later down the track, I applied for the internship program and was successful and got a position to do a summer internship between my after my Honours and before my PhD.
So there's a video of you talking about your time at the internship.
And one of the things you mentioned is that you were the only biologist in the room and all the rest were engineers or IT people.
I guess was that daunting when you first rocked up thinking, did I get into the right place or am I going to be okay here?
And with that kind of knowledge in the room, that was different.
Absolutely.
It did feel very daunting to me.
The crash course that they ran for the first week for a lot of the interns was sort of looking at
skills they already had and just refining them.
And for me, a lot of that was brand new learning.
So it was very daunting at first when we went around the room and each spoke about our background.
And I did, I think, look over at Mark for one minute and think: “Oh no, what did I apply for and how did I get in?”
But what was really cool was that most of the projects were actually biology-based, even though they weren't biologists doing them.
So there was certainly, even at that point, a real need for biology questions being done in a
more computational and IT-centred expertise.
A lot of people
who go into using HPC, for example, have the backgrounds that are not in computing, first off, it's usually in biology or chemistry or these sorts of things.
So having to learn programming or command line work, again, this must be pretty daunting if it's not your cup of tea or if it's not what you want to do.
But I guess when you started to learn that part of it,
when did you start to realise that it's opening up, I guess, doors in terms of your research?
I guess as well I had finished my Honours by that point.
So I already had known firsthand some of the bottlenecks that can come from analysing data locally and things like that.
Once we got that crash course of coding and we got a week to go back to our projects and delve into some use cases and some examples that we wanted to gather the projects, it was sort of at that stage that I thought making my timeline of the twelve weeks, how much we
can really accomplish if it's full time just working on the coding as opposed to what is a usual life scientist workflow, which is you divide up that time between the field and the lab and you do a bit of data analysis at the very end of your project.
So I think quite early on maybe for me, knowing that it was really going to be beneficial and there would be a time where the skills don't seem directly relevant to the outcome, but just
hoping that it's just a bit of a painful time and to ride it out.
So those sorts of things where you're able to ask different questions or expand your work, did you feel like after that this is something you would try to continue using, bringing HPC into your workflow essentially?
And I think as well, again, as a background in a life scientist, the rate of sequencing, the cost is going down continuously.
So more and more, you're needing to do more samples and analyse it in a different way, in another way, because the cost is going down so quickly.
So trying to keep up with the latest research means that it's inevitable that you're going to have a huge number of samples now.
And I think now in definitely in genomics, the bottleneck is the analysis.
we're getting to the stage where it is quite cheap to get some data, good quality data quite quickly from samples, and it's the analysis that is the bottleneck at the moment.
What makes a tick tick?
So again, we take it for granted sometimes, but the first thing I would want to say is that ticks are arachnids, actually.
They're closely related to spiders, and they're in that group, so they're not a type of insect, is the first thing that often gets misquoted in the media.
But ticks can also transmit, they can actually make you allergic to meat.
So ticks in Australia, we don't luckily have them on the West Coast, but on the East Coast, there's a mammalian meat allergy syndrome that can, after just one bite off a paralysis tick, it can make the person allergic to meat and meat products as well.
Really?
Like forever or for as long periods of time?
It's essentially lifelong, but evidence has shown that after ten years, it does seem to drop off.
But that evidence is still new and the syndrome was described actually by an Australian immunologist over in the Northern Beaches, Michelle Van Ewen, who is an amazing woman that put together the pieces of the puzzle after many years, where she saw her patients getting this delayed allergy to meat products and concluded that it was from a tick bite.
So that's something that is increasing, not just in Australia, but also globally, as since being described in Northern Hemisphere and other parts of the world.
We're so used to thinking, it's probably because of this or it's just an unknown thing.
We're starting to learn a lot more.
And as you've just shown, like these ticks can potentially cause people to not be able to eat meat anymore is quite a discovery, really.
Absolutely.
And I think as well, there's still new things to describe is the moral of that story is that even though these ticks are not new, they're native ticks that have been in Australia for
thousands, probably millions of years, but it's only a newly described syndrome that even in, like you said, the last 12 years, we've got more and more information.
And now they're doing work about the best way to remove the tick to avoid that type of syndrome occurring.
And the research is still at its infancy as well.
How many different species have you gotten to work with, I guess, over the years?
Do you have like a
round number of how many?
So there's about 74 species of tick in Australia and through our lab we've probably seen the vast majority of them because a lot of them, our wildlife have the largest diversity of ticks present and we're fortunate enough to have collaborators across Australia to collect ticks on our behalf or we have projects that mean that we go out ourselves and collect them.
So I would say, I haven't put a number on it before, but I would say probably at least about 60 have come through our lab to some extent.
Have you ever had a tick bite?
Absolutely.
Unfortunately, it comes a bit with the job.
Yeah, both on the East Coast and the West Coast and anywhere that we're unfortunately collecting ticks.
You try and suit up as much as you can, but often our aim is to go out and collect the ticks themselves.
So unfortunately, a tick bite is inevitable, but we come prepared with the right tools to get them off quickly and tick checks and as much PPE as we can.
I was always curious if you're working with ticks, it's the same with you think about working with bees or any other kind of, you know, insect or animal that has some danger to them, thinking like “Oh, please don't bite me, please don't bite me.”
Oh, it's okay.
It's fine.
Well, hopefully
you haven't gone off meat at all.
Or are you vegetarian?
I mean, that's also a thing.
That's also a thing, but that was that was definitely my first thought when I got bitten by a paralysis tick over on the East Coast is because it transmits so quickly as well that it can be can be within hours.
So, really that quick?
Yes, and often, yeah, we can't wear the insect repellents or the repellents for the ticks because we're trying to collect them ourselves, so we have to use other methods.
It'd be a long, night trying to catch ticks if they could smell DEET from a million miles away.
Just like our native animals and our native plants are really unique to Australia, so are the ticks themselves and importantly the microbes that they carry.
So often in previous years gone by and decades ago, they were looking for agents of disease that were described in Northern Hemisphere and seeing if they're present in Australia.
But what our research has shown is it's most likely not that simple and what we have may be something that is different to what is described in Northern Hemisphere.
So we're really looking
for something unknown.
And that can be really hard to do.
And that's why we need the unbiased power of genomic sequencing combined with HPC for our analysis as well.
I'd imagine it's very hard to work with genomes now being that there's so many now that have been collected, I guess.
How is it to work in that space nowadays, given that there's such a large data set, I guess.
Previously you could use a curated database and you could do your analysis locally because there was only a few options to choose from in terms of matching your sequences.
But now to get the best results, you want to match it to every single thing that has been either named or not named.
And to do that requires a lot of storage.
So not just the HPC, but also the compute and the storage that goes with that as well is really important.
And that's something that Pawsey is really ideal for, having that side-by-side power to do that analysis as opposed to if you were to try and substitute that for cloud-based or for other parallel programs as well.
So if you had something like cloud or some other system, would it like dramatically slow down, I guess, the work?
You can do parts of the pipeline.
So there's perhaps unlike with astronomy data where there's sort of a real bottleneck
At one at that acquiring the image stage and converting that to some kind of data that the computer can interpret for genomics, there's quite a few steps along the way, and some of those can be done in the cloud compute or can balloon out when you need them, but there's certain parts of.
the algorithm like when you want to match to all the other sequences available that really rely on having some powerful HPC to do the best job.
Is there a particular space in which there's still so many unknowns that I guess
you and your team really just want to get into and either you don't have enough time or you're just hoping in the future you're able to dive into?
Absolutely.
I would say during my PhD it's a bit of a frustrating curse that everywhere you look there's something new because the world of taxonomy is painstakingly slow to be able to describe new species and validate them and publish them is
the first step before you can then make that leap into is this causing disease and what is the ecological niche of that organism.
And everywhere we look even in the black rat that has been here for
been introduced previously, even that has our native black rats that have come from overseas, even they have something that is different to what is described in the Northern Hemisphere.
Just by the time, even in the couple 100 years that they've been present, they have evolved a slightly different strain of these organisms.
That means we have to go back and describe that first and before we can do anything else.
It must be really hard to
not challenge, but I guess add to what I guess the Northern Hemisphere knows about black rats and go, well, these ones in particular in the last 100 years have somehow changed to get to this point.
And then having them to accept that and go, well, sure, but there's also these, we know black rats are over here and this is how they are there.
Do you ever find that when you introduce something new to, I guess, the known species like the black rat, that it's
that it's hard to sell to the idea that they are different?
Are there certain aspects of it that it's really hard still to get people's heads around the differences between...
the same animal in a different place or--
It is and it is as well because we are so far behind.
So I would say in the world of tick-borne diseases we're at least sort of 50 to 60 years behind the northern hemisphere in terms of when they started describing a Lyme-like illness in the Northern Hemisphere and it's the same probably across the board.
And so when you go to publish papers or they get peer reviewed and the reviewers are from the Northern Hemisphere they're often asking you to do the next experiments and you have to just remind them that you know they are coming but
sometimes science can be painfully slow.
That does happen, but it's a process where you sort of put one brick at a time on the foundations in the hope that maybe it's not even you, maybe it's those that come after you finish off building the house.
I guess a project you're working on at the moment that you're really excited about?
So the latest project that we've been working on is a NHMRC-funded project into the debilitating symptom complexes attributed to ticks.
It has a very catchy name, abbreviated to DISCAT, but essentially it's trying to work out that now we sort of have an idea of what's present in the ticks and in the wildlife, can we find something in the humans that may be causing that illness?
Right.
So it's just learning about the human chemistry related to what ticks can bring over.
So trying to find out if, because of how we're structured, we might receive one disease and not get another kind of thing.
Yeah, absolutely.
And as well, just because we find something in wildlife that doesn't cause disease doesn't mean that when it jumps to humans, it will
be benign.
So we know that when things jump host, they can also change their pathogenicity.
So it's about untangling those intricacies now in the focusing on the human space.
Have you found anything recently that has been really surprising for you and your team?
What we've done so far is we have worked with a bunch of collaborators to get a suite of data looking at the viruses present, the bacteria, the parasites, but then also the
So looking at their transcriptome profile, looking at small RNA molecules, looking at their other inflammatory profiles, and then to pull all of that together to see if we can find some intricacies with
matching up a certain microbe to a certain disease phenotype and trying to see what that looks like over the course of a year for a person.
So we're following these people over the course of a year after they've been bitten by ticks.
Right.
So we're looking essentially first at the microbes.
So what microbes are present?
And that may be viruses, that could be bacteria, or that could be other parasitic parasites, things like malaria that don't fit into the other
virus or a bacteria.
So we're looking at the microbes present and then we're looking at what the immune response does for that person.
So we're looking at the person at the time of the tick bite and then we're following them over four time points up until a year to see how they respond acutely just after a tick bite up to a week or three months and then what that person looks like after a year.
Have they recovered or are they displaying any of the same symptoms?
And can we tie those symptoms to a persisting microbe that is present?
Like long term, I guess, that kind of data, is that relatively new for you guys?
It is, and it's also new for the world.
Really, no one in the world has done a study that has been a prospective longitudinal study for people that are acutely bitten by ticks.
There's, like I said, there's been a lot of studies in the Northern Hemisphere looking at people that they know are sick.
and trying to find out the agent that way.
But what we have found is that in Australia, the population numbers mean that we can't rely on a big outbreak occurring because we just don't have the numbers.
And by the time we get that big outbreak, that means that there's millions of people that are going to be infected.
So we need to do something pre-emptively in these cohorts, which means that they're, I guess, smaller in comparison to the normal studies done in the northern hemisphere.
And we're trying to overcome that barrier with doing multiple-omic strategies by looking at the different types of RNA molecules, DNA molecules, but also proteins as well that have been translated in the individual.
Getting to do that kind of new study presents itself with its
just as much excitement as it is going to be, what do we do afterwards and how is it going to either be replicated or worked on from there, I guess, which is really cool.
So in particular with this type of data set, we're looking at terabytes of data per individual patient.
And to analyse that accurately, we want to compare that against every single sequence that has been generated, known or unknown, that's available publicly.
And to do that, we need HPC.
And how long does it take to usually go through all those genomes, I guess?
So an individual person's data set can actually be analysed in just a couple of days when we use resources
like HPC coupled with the storage that Pawsey can provide, as opposed to what would take weeks or perhaps months for to do locally.
And also something that we can't replicate locally is that annotation, comparing our sequences against everything else is just something that is not possible to do locally.
Yeah, I would imagine, yeah, your laptops probably won't be able to do go through every single genome in existence.
I guess what I'm really excited to start working on more next year is some training and some teaching.
A little bit different, but I think I'm at the point now where just like in 2017 where it was mainly IT and computational people doing the analysis, that still is largely the case.
That happens a lot.
So I think they, like I said, sequencing is getting to a point where it's so accessible that people are spending less time in the lab, even if you are a traditional lab scientist.
And that's where it is at now, is understanding the analysis and that's where the time
is best spent.
But unfortunately, with the undergraduate degrees to date, they haven't spent the time learning, they haven't spent the time teaching those sort of skills.
So I'm really excited actually next year to start some both undergraduate but postgraduate training and teaching in the bioinformatics space.
That'd be great to get to showcase, I guess, the power that goes behind just having that skill set, I guess.
And having to go through that yourself and then getting to share that with new students and why it's important for you to look at this as a really helpful tool in your tool bag.
Exactly, and I think I've been lucky that I have seen the power that HPC or having access to HPC can provide.
And my strategy is to showcase that power early on in the semester.
So getting them, week one, week two, generating those amazing figures and then going back and learning the nitty gritty code and the other bits of
tools that are needed to get to that point.
But I think as a life scientist turned computational biologist, I have a unique aspect to teach.
Yeah, because you've done teaching before, right?
You do lectures and things like that before, right?
Yeah, absolutely.
And it's mainly been in the more in the life science space, so teaching those fieldwork techniques or teaching those laboratory techniques.
But I think more so now in the data analysis space as someone that can
the biological aspect of that data, that's where things are headed next year.
Is there anything you miss about, do you get to go out to the field as much or does it feel like you're less getting to go and collect now that you're working with people?
Definitely less.
Although having said that, there was still quite a lot of collection.
I would say the people are just as
just as resourceful, intensive to sample as the animals in a lot of respects, convincing them to partake in the study and getting that information.
So I would say I still probably spent a lot of time or just as much time collecting from the humans and interacting with the study participants, which is always nice to be able to explain to them the study and why they're
why we're doing what we're doing.
And I think like with most life scientists, you go through phases where you're in the lab a lot or you're analysing data a lot.
But I would say overall, I am generally now the person that people come to with their data queries and I do spend more time in front of a command line terminal.
I feel like maybe the DEET of humans are emails.
So it's probably hard when you send an email to get someone's attention.
It's more like: “Oh, no, I'm too busy.
Leave me alone.”
But if it gets to sit down and tell them like the reason why we're
following up, it's hard to get people on board at first because I guess they don't see the bigger picture of why you're collecting all that data in the first place.
Absolutely.
So I think, yeah, education is the key to get them to participate, but we're very lucky with our cohorts of patients that were very willing to volunteer their samples and their time trial study.
The amount of data that comes in for, like, in biology and just in general, I guess, is there
somewhere you see HPC going in the future in terms of dealing with all this extra data and is there anything that you see in the future that might be good or bad for it?
So there's some really exciting advances in genomics in particular with real-time sequencing now.
So now these devices can read individual base pairs essentially, those A, T, C's and G's in real time.
And we can collect that data.
So I think the next thing really is to then now start analysing that data in real time.
I guess there's no point collecting it in real time if we don't analyse it in real time.
So that's really the next step that I think HPC can be really powerful for.
The data they collect is huge and it really needs some humans to train that and learn the algorithms that is potentially something that could be helpful in the future and I think definitely needed.
The next step after we fine tune the sequencing platform.
Is there something that you know about collecting that data in real time that you feel like diminishes over time that you need to have that kind of real time collection and analysis at the same time?
So an example is sometimes what these platforms can do is they can target certain fragments of DNA that you're interested in.
And they do that by reading the first maybe 10 or 15 base pairs and decide if they should keep reading that fragment of DNA.
Is that what you want?
Should I keep reading this or should I discard it?
So to analyse that, so you need to analyse that in real time to be able to do that.
And that's something that is being done more and more with this use of targeted metagenomics is what it's sort of referred to, where we're looking at a essentially a needle in the haystack and we're trying to find that needle and it wants to know if it should waste time sequencing that particular strand or if that's not quite what it's looking for.
It's really fascinating having to find.
like a particular type.
I guess when people think about DNA sequencing, they just think about long sequence, the curly sequence, and that's your genome, but it's harder to think you need to take fragments and figure out whether it's worth continuingly studying, I guess, that string.
Absolutely.
So that curly sort of big long fragment is, I guess, the nice pretty data you get at the end, essentially.
To get there, you have to break up your DNA into fragments often, and you have to put those fragments back together computationally to get that pretty diagram.
And to do that, we can optimize different workflows.
And yeah, one of these real-time sequencing platforms can in real time reject or accept different fragments depending on what you're looking for.
To me, it sounds really fascinating having to make those decisions.
I guess you've got programs that are able to, I guess, figure that out real time.
Is there something within a sequence that once you start looking at it, you don't need?
Like you're saying that sometimes they'll look into it and then it doesn't need to.
What kind of things in DNA cause you to stop?
looking, I guess, in that strain.
So essentially you can have a blood sample.
So a human blood sample is a good case where you're looking maybe for a bacteria in that blood sample.
And so in that blood sample you probably have about 95% of the DNA is going to be human.
And you don't want to sequence that human DNA because you're not interested in that.
You want to know what the bacteria are.
So it will start reading a piece of DNA and go: “This looks like human.
I'm not going to keep reading this piece of DNA.
I'm going to chuck it out.”
And I'm going to try again now and get a bacterial read and sequence that.
So that's the kind of example where it can reject certain fragments of DNA.
Super cool.
I didn't know you were able to do that.
It is.
Very cool and it's very novel and the algorithms are still very much in the infancy and at that sort of alpha sort of development stage.
But the potential, I guess, to be able to pull those things out and find out in say a blood sample, getting to quickly detect things, I guess, especially when it comes to infectious diseases and things like that can really help when you have a short amount of time before.
a human might have worse reactions and you need to find out what it is really quickly.
Absolutely.
The traditional way is to take a blood sample and put it in a bunch of different selective culture medias that may or may not target what you want and often take days to grow.
And by the time you get those results back, it may be too late to intervene.
So it can be really powerful to just get that human blood sample, run it through a sequencer and get those results in real time.
Do you feel like
once that starts to evolve, there's going to be a lot more movement or excitement in terms of focusing on that as a medical sense, trying to find out what's going on with someone's body?
Absolutely.
I don't think it's made it into the clinical phase just yet, but there are certainly clinical labs that are testing and researching these kinds of developments along with academia and our research group as well, all playing a role in looking for their own certain or their own favourite pathogens in a human blood sample, getting all that data together
to optimise a clinical workflow is the ultimate goal.
Is there anything that you did in the sort of the animal world that could also benefit from that kind of research going into a blood sample of like a horse or a dog and being able to find out these strands of bacteria or viruses that are in their system?
Definitely.
And particularly in resource poor setting sometimes.
So sometimes if you are sampling a sick dog in a rural or remote community, there's no time or no resources maybe to go back and take that sample to the lab to culture it up.
So also developing these so that they can be done in a low resource setting is also really exciting for not just for humans, but for other animals as well.
That's huge.
That's a huge development.
Yeah.
When we talk about having to
learn a new field or new space, usually you think it's really directed to, it's very close to what you're already studying.
So if it's life science, you think, okay, we're going from a tick to a flea, or that's something that you already kind of knew a bit of the structure, so it's not as hard to adapt.
But when it comes to something like programming, which it's, you know, very much a computer heavy thing to learn, it must come up a lot as
not a pain point as such, but it might make people fear having to go into that space, I guess.
Is there anything you can share that I guess might alleviate that?
What I often tell people to approach the sort of computational side is to think of it at least to begin with, just like a tool in your toolkit.
So you don't have to know how to build a computer to be able to log on and answer your emails and do what you need to do.
And it's the same with HPC.
Other than some fundamental skills in coding, you don't need to memorize every step or every command off by heart.
As long as you know where to find the information that you need, you can begin by just
being a user.
And I encourage that most life scientists end up just being, I say just, but end up being users of it.
And there's absolutely nothing wrong with that.
The people that develop the software need users for it to work.
And so you can just be that user.
I think when people often think about HPC, they think about developers and developing new software.
And that certainly can be where some computational biologists end up is doing that.
But it's not necessary.
You can
user stage.
And I really encourage new students or new researchers to approach it in that kind of way.
I guess it's when they see results that it makes it easier for them to go, okay, I should probably get behind this because there's going to be a huge benefit if I don't or if I do.
Thank you so much for sitting down and chatting to us.
I know there's so much more we could probably pick into what you're working on, but it's really cool to get to see.
I guess the potentials of what you're working on that can go into many different fields like medical and that sort of thing as well.
So yeah, thanks so much for your time.
Thank you.
Thanks for having me.
Thanks for listening to our chat with Siobhan.
There's a lot more stories to come and we can't wait to share.
If you are interested in what supercomputing can do to research, visit our website for more stories like this.
Until then, we'll talk soon.