Episode 5 - Genomic Futures: Impact

Show notes

Speakers:

• Dennis Chopera, Senior Programme Manager at the Leadership for African Research Networks (LEARN R&D);
• Olivia Allen, Head of Strategy at Wellcome Sanger Institute;
• Dana Cortade, Project Manager at the Align Foundation;
• Tariq Khokarr, Head of Data for Science and Health at Wellcome;
• Sophie Gilbert, Programme Lead at the Geneva Science and Diplomacy Anticipator (GESDA);
• Alexandra Canet, science communicator and producer of the Genomics Futures podcasts;  Matt Hurles, Director of the Wellcome Sanger Institute.

Episode description:

In this fifth episode, we looked at the themes that came out from the Novel Ways of Achieving Impact workshop, in which themes on lab automation for genomics research, equitable life sciences ecosystems and the future role of wearables. We spoke with Professor Matt Hurles and main instigator of the Genomics Futures workshops, who chaired this gathering, alongside Dr Tariq Khokhar.

With Dr Dana Cortade, we spoke about AI and changes to data generation and data integration. We then looked at what an equitable life sciences ecosystem would look like in 2050 with Dr Denis Chopera and Dr Sophie Gilbert.

Mentioned in the episode:

• GROQ-seq platform – A project managed by the Align foundation
• UK Biobank – a long-term prospective biobank study in the United Kingdom (UK) that houses the de-identified biological samples and health-related data of half a million people.
• Digital twin< – a virtual replica of a biological entity or process. It can use real-time data to simulate, predict and optimise outcomes.
• MLW – The Malawi-Liverpool Wellcome Trust Clinical Research Programme (MLW) is built around excellent laboratories, strategically located in the largest hospital in Malawi, Queen Elizabeth Central Hospital closely linked with the community and an integral part of the medical school.
• Africa CDC – The Africa Centres for Disease Control and Prevention is a public health agency of the African Union to support the public health initiatives of member states and strengthen the capacity of their health institutions to deal with disease threats.
• NIST – The National Institute of Standards and Technology (NIST) is an agency of the United States Department of Commerce whose mission is to promote American innovation and industrial competitiveness.
• LDL – Low-density lipoprotein cholesterol, often referred to as “bad cholesterol” because it can build up in arteries if levels are too high increasing the risk of health issues.

Transcript

Dennis Chopera 00:00
For me, in an ideal world in 2050 I think genomics will be such a powerful, accessible tool for health equity as well as social transformation. I think it will enable precision medicine, including early detection of diseases as well as personalised therapies that I’d select specifically to accommodate the genetic diversity of different populations.

Olivia Allen  00:29
What will the future of genomics look like in 2050. Welcome to podcast five of the genomics futures series. I’m Olivia Allen, Head of strategy at the Wellcome Sanger Institute, and your narrator throughout this podcast. Today, we’ll be exploring the insights and topics that came out of the ‘Novel Ways to Achieve and Deliver Impact’ workshop. This workshop looked at various themes, including AI and changes to data generation, wearable technologies and new organisational models for the advancement of life sciences research.

Dana Cortade 00:57
I think that the implications of what we can achieve will be so vast that it’s kind of hard to cover them all, but in particular, I think that our knowledge of how the natural world around us works will reach a point where we’re able to interact with it on a totally new playing field.

Olivia Allen 01:18
In this episode, my colleague Alexandra Canet, co producer of the podcast series, explored these themes with several attendees. First to introduce the themes, Alex spoke with the workshop organisers, Professor Matt Hurles, Welcome Sanger Institute director and Dr Tariq Khokhar, Head of Data for Science and Health from Wellcome. We asked them to reflect on the 25 year exercise and how the outcomes from the workshops might impact their respective institutions.

Tariq Khokhar 01:43
That’s a good question. As a global philanthropic foundation, we can’t really just respond to science as it unfolds. We really have to be in a proactive position to shaping the ecosystem that science needs, and try to spot kernels of ideas that could become entire fields in the future. So from this sort of 25 year exercise, it has both sort of encouraged us to look back at ideas that may have seemed a little bit far fetched when they were first born and see how they’ve played out now. But it’s also allowed us to think now, what are the kind of kernels of possibility that could turn into something impressive in the future.

Alexandra Canet 02:24
You mentioned there, kernels of ideas, any that might have come up from the workshops that come to mind.

Tariq Khokhar 02:30
I don’t know how long we have in this podcast, but there’s a lot. I think the biggest, the biggest idea, is really that, for me, data is no longer sort of downstream of discovery. It’s really the substrate for it, and we’ve known that for the last 15-20 years, if you like. I mean, the entire premise of things like Sanger, the UK Biobank, are really the data is the sort of foundation of these enterprises, and research and discovery sort of flows from them. But it’s important to remember that has always been the way that science and discovery has been viewed.

Matt Hurles 03:04
I think the exercise of thinking over 25 years is not something that we tend to do in our kind of research communities, and it does require thinking about things in a different way, and it does start stimulating, as Tariq says, that kind of looking backwards and kind of thinking about 25 years ago and what somebody 25 years in the future might be asking of you. But I think one thing that really kind of struck me is just the interdependencies between the different themes. AI and all its myriad possibilities intersect very much with equity and the equity of science going forwards, and also with organisational models and how we structure ourselves as scientific communities, as scientific organisations. And although we kind of picked these three themes apart, they keep colliding back together again, both in this workshop and in the other workshops.

Alexandra Canet 03:58
There were three main themes that came out of this workshop that I’d like us to explore today. The first one is artificial intelligence and changes to data generation and integration. Which are the biggest changes we might see in 25 years time.

Matt Hurles  04:13
So I think maybe starting with one of the avenues that we discussed quite a lot was around the potential of a much greater automation in the laboratory. Especially around the kind of modelling of biology in model systems like cellular systems. And that presents opportunities for generating much more robust data sets at much greater scales, but also more fundamentally, it can bring the opportunity of new ways of doing data generations. So rather than humans designing experiments, potentially AI designing experiments iteratively, and what’s sometimes being called the lab in the loop. I think there’s also the prospect that some of the tasks that researchers do will become taken on by AI. It’s hard to predict exactly what those tasks will be but it’s quite easy to predict that the role of a researcher could look very different in 2050 from how it does today. And then I think there’s the third thing, which is the idea that will generate these massive AI models that will enable us to simulate biology, sometimes called a virtual cell or a digital twin, but the many more experiments will be conducted in that in silico virtual space rather than laboratory experiments.

Tariq Khokhar 05:30
Thanks, Matt. No, I definitely agree with those three, and maybe to reflect on one and add a couple more on the automation feedback loops and that fundamental piece around data generation. I do worry sometimes that folks see kind of modern innovations in AI and think, wow, AI is just done now. We’ve got Chat GPT, we’ve got all these kinds of cool tools, but fundamentally, these systems can’t reason about domains at which they have no knowledge, and that knowledge is represented by data. And there are huge swathes of the biological universe where we just don’t have the data and telemetry about them, so that data piece is going to be incredibly important if we want to leverage these new AI capabilities.

Alexandra Canet 06:13
Building up on that, I was looking at my notes when I was preparing for this podcast, and on this theme, I had written many, many times, the word wearable. So what’s the role of wearables in all of this?

Matt Hurles 06:27
Well, this is definitely much more conjectural, because we’re requiring really substantial technology advancements to make this work. But essentially all of genomics at the moment is predicated on taking samples from people and that has certain limitations associated with it. You can only take them so often, the people that have the time and are able to financially contribute are often a biased subset of the population, and we’ve seen that with many cohorts. So the prospect of having something that’s wearable, that requires less of its participants, but also has a much denser longitudinal kind of following by molecular measurement of individuals. So that kind of prospect is really, really exciting, and that’s obviously quite science fiction. I mean, we’re slightly in the realm of Star Trek here, but I think it’s well worth thinking about, because we’re already seeing things like, you know, glucose monitors, things that people are wearing, that are measuring glucose and then, if you’re diabetic, dispensing the right amount of insulin. So, not just measuring, but also intervening in a tailored way.

Alexandra Canet 07:38
The second theme that came out was looking at institutional models and how they’re going to evolve over the next 25 years. There was a lot of talk in increasing the diversity of organisations, for example, or who should be setting the agenda. Again, can you paint a picture of what all of this could look like in 25 years time?

Tariq Khokhar 07:59
For me, that future has to be engineered and not hoped for. It’s not going to happen unless you’re deliberately pursuing it. There’s a few aspects of that I’d focus on. So one is fundamentally about a global participation and global inclusion in science, and that means making sure we’re able to work not just across geographic borders, but also across, probably disciplines, and increasingly, across the machine and human boundaries as well. So there also needs to be a shift in how we think about the flow of talent. At the moment, we will often have, for example, in the UK, you might have a tension between talent going to industry versus talent in academia. Globally, you may have a bit of a tension between talent leading lower income countries coming to wealthier countries and potentially causing some sort of brain drain, but there’s lots of ways to think about that very differently. What would it mean to have more porosity between industry and academia. What would it mean to create centres and environments in middle income countries where those folks can have great careers and thrive, et cetera. And I guess the last thing is, fundamentally, it’s about building long term capability and capacity in more places, so that the folks who are ultimately benefiting from the science are both closer to the science, are participating in the science and are ultimately shaping its trajectory and future.

Matt Hurles 09:24
Yeah, so I guess on that more kind of equity note that Tariq was talking about, one conversation I had over and over again in the workshops, was that sense of the brain drain currently, of the you train people locally in a lower income environment, they become very desirable to recruit into higher income environments. And the ability to sustain those kinds of communities is really, really hard, and I hope that by 2050 we’ll have a greater diversity of organisational models and that certain types of projects, and especially the projects that maybe generate the very large data resources that are useful for AI, you know, maybe those require different kinds of teams, more multidisciplinary teams, bigger teams, maybe longer projects, and that the university PI model has been incredibly productive works very well for certain types of science, especially technology development and mechanistic biology. But there’s other kinds of scientific questions, other things that we’d want as a society to address where actually we’re going to need different models, and actually we’re kind of in an exciting and maybe pre-Cambrian explosion phase, where lots of different organisation models can be explored over the next 5-10 years, and by 25 years time, we will have a greater diversity and a better sense of what kinds of scientific problems could be tackled by what kinds of organisations.

Alexandra Canet 10:53
Fantastic. This leads us really well into the third vision that came out during the workshop that was about making life sciences, the field, more equitable in 25 years time. I think you’ve touched upon a lot of different ways that this could happen. But again, what will equitable look like?

Matt Hurles  11:12
Yeah, a couple more thoughts from and especially formed by some of the more recent conversations I’ve had at the latest workshop. One was around not underestimating just how inequitable it is currently. So it costs vastly more to generate the same data in most low- and middle-income countries than it does in a higher income country. It’s not that it costs the same, it costs vastly more. We have collaborations in Africa where it costs them seven times as much to generate the same quantity of data as it does us. So that inequity of cost is not even a level playing field in terms of the cost per unit of data.

Tariq Khokhar 11:49
Just to add on that briefly, I mean, that’s exactly our experience as well. And the challenge is going back to your macro lens, it’s like, how much can an organisation like Wellcome help with that, and how much is a broader kind of development question. As you know, we support the MLW Centre in Malawi, and when I was there, exactly to your point, they asked researchers, what are the things that are constraining you. And one of them was actually an internet connectivity. If you want to use a cloud based genetic analysis platform and you want to upload a three gig file on a network connection that is worse than a mobile phone in the UK was five years ago. That’s not a sort of great experience. And equally, things like cold chains for moving samples during the pandemic, when we were involved in supporting some of the standing up of pathogen genomic sequencing facilities. I remember talking to a team trying to get samples to the Africa CDC, and they were basically having to ship in dry ice from Europe as the medium to pack samples, to drive it across the country, because no one wanted to put it in a plane to totally agree. I think it is remarkably inequitable in ways that you probably can’t even imagine.

Olivia Allen 13:03
Dr Dana Cortade, Project Manager at the Align Foundation, was a workshop participant. In today’s podcast, she explores one of the main themes, AI and changes to data generation and data integration. First, we asked her about her views on the future of genomics.

Dana Cortade 13:18
I think that the implications of what we can achieve will be so vast that it’s kind of hard to cover them all talking about the utopian version of what could be. There’s this transition of in the 90s and before having really heavy plastics and lots of things that were highly chemical and causing some problems. And I think that by the time you get to 2050 as just like a very consumer level example, you’ll be able to have products, and you know, materials that you interact with on an everyday basis, that are safe and that you feel safe with, because you know that they’re made from a source that’s not dangerous for you, it doesn’t have any of these crazy chemicals that sit around forever that we just realised, oops, we put that everywhere, those things will be cleaned up because we’ll have engineered systems that can detect them and then alert us, we’ll have engineered, hopefully different enzymes or different things that can break them down or make them useful.

Alexandra Canet 14:14
Fantastic. So at the workshop you attended, automation and genomics at scale came out quite a bit. I think it’d be useful to understand what we are able to do now, and what we hope to be able to do in 2050.

Dana Cortade 14:28
Yeah, so my work is more on the protein engineering side of things, and getting into synthetic biology world, and in that kind of lens, naturally we’ve done before a lot of in vitro assays, which is where you take some sort of protein or enzyme or something out of its cellular context, and you put it in a plate, and then you test it for its different properties, whether it works well or not well, and you engineer new systems based on those results. And we could do that maybe, like hundreds of proteins, maybe even 1000s of proteins, but that takes a very long time to be able to express and then purify these, so we make sure we’re testing only the thing that we’re interested in, and then actually making an assay just specifically for that kind of protein. So if you think about changing the genetic structure of an organism, or even just the gene to express the protein. There’s a whole pipeline after that where you actually have to do these tests, and they take tonnes of man hours and tonnes and tonnes of money. It’s very resource heavy, but already we’re starting to see really cool high throughput assays coming out. And the difference that makes it high throughput is that we’re no longer measuring 96 enzymes or less at a time on a single plate. We can measure hundreds of 1000s of different enzymes or hundreds of 1000s different proteins at the same time in the same well and in a cellular context. So we’re actually seeing them working more similarly to how they would work in their natural environment, versus necessarily taking them out and using them like we do in certain products, but keeping them in their context to see all of their aspects. So I think that’s a huge thing, being able to have high throughput automation to enable lots of experiments, because if we’re able to have lots of data on lots of experiments, then we start getting in the world where we can apply machine learning techniques and AI techniques.

And then the other part of that that we’re missing right now is the idea of reproducibility in biology. There’s a lot of other fields who have made these huge leaps, making large data sets and then training different AI and ML models on them, but usually those data sets that they’re being collected are coming from industries that have highly standardised components in them, like the electronics industry, but biology is so complicated that we really need to crack down on reproducibility. And using automated methods really helps with reproducibility, and we still have a really long way to go in ensuring that we can have people at different sides of the world doing the same experiment and getting the same results, but we’re actually seeing that happen. I actually, this interview came at such a cool time because I lead a group for the Align Foundation that’s called the GROQ-seq group. It’s growth based quantitative sequencing, and it’s looking at the relationship between protein sequence and function. And we make these really high throughput assays, and so we have a transcription factor assay that you can run anywhere from 100,000 to 500,000 variants in a single well. And that works really great in the hands of the team we’re collaborating with at NIST in Washington, DC, in the living measurement systems foundry. But if they do it, and they are the only ones that do it, like, what does that really say for the context of biology. And that was a big question. So we’ve been working on onboarding that at the damp lab at Boston University, and this morning, we just got our first huge results back. And I wish I could show a picture in this podcast, but the results are incredible. If you imagine a correlation plot and you just have a line going perfectly 45 degrees, all of our calibration variants are basically sitting on that line between the two sites. And there’s just such a huge win, not just for us, because everything we do is open science, but even being able to see that this is something that is pretty complicated, a series of protocols being replicated on different sides of the United States, totally different teams buying things at different times.

Alexandra Canet 18:42
That’s excellent news. Yes. And going back before we talked about reproducibility, you were talking about all these experiments and how the data was going to enable predictions. But what will the general population gain from all of this? Can you give us some practical examples?

Dana Cortade 19:01
100% there are so many good examples of this. So from a personal level, there is a possibility that you could have a type of medicine where you’re being cared for before something bad happens to you. So for instance, people who have genes that impact their body’s ability to reduce or remove cholesterol from their system, and there’s a condition called familial hypercholesterolemia. Basically, it just means people with that don’t have the ability to remove the nasty type of cholesterol, LDL, from their body, so whether or not they eat well or exercise or do anything, they have this huge buildup of LDL within their systems, and that’s exactly the type of thing that gives you a heart attack in your 30s or 40s, and so it just hits out of nowhere. If you didn’t realise it’s in your family, then there’s nothing you can do, and there’s not a tonne of exterior showing symptoms to show you what’s going to happen and that kind of thing is something that is actually a genetic marker. That if, when you’re born, you do a screen, a comprehensive screen of genetic markers, or maybe we have information from your parents, and one of them had it, or both of them have it, then we can do predictive medicine and screen you for it or even in the cohort of people, eventually not even having to screen every person and sequence every person’s DNA, but knowing that here is the family lineage and then another one that I think a lot of people would be able to relate to is pandemic and epidemic awareness and preparedness. So that’s the idea that we all lived through covid 19, and that was something that hit, and then we had to react afterwards to it. And right now, a lot of people are investing in trying to make data sets or make models that can be predictive of, you know, if I put in what a sequence of a virus is, or like what what its backbone, or what it’s made up of, I can also have this tool that, if I put that information in, like, this is what it looks like, or this is what it’s made of, it can tell me, well, this is a good binder, that will work to stop it, or this is a good type of medicine to be able to take to treat it, or something like that. And so right now, what we have to do is, we are just getting to the beginning of being able to do that. But the idea is that in the future, as soon as something happens, we can react to it quickly and also have a very rapid solution to be able to test.

Alexandra Canet 21:32
These are fabulous examples. Thank you. I’d also like to get your thoughts on what was missing from the conversation during the workshop. Is there anything else we should be thinking about?

Dana Cortade 21:42
I guess one of my biggest takeaways from the workshop itself that I don’t hear talked about in other venues that I’m typically in, is the idea of really reflecting on what is best for funding and for building centred and investing in different regions having access to different kinds of sequencing equipment or equipment to be able to do different types of experimentation. Because everything that I work on and everything I’ve just said is all around, you know, wanting to have reproducibility, and the reason that we need that as well is that if you have a single place that does something, what happens if something bad happens to that single place, or it loses funding, or it doesn’t have that equipment anymore, so there’s this risk of having a single site, and what we want to do is have redundancy. So we have lots of people doing lots of things everywhere to make us stronger as a global community. And we think about that in terms of doing experiments at different places, and wanting to have protocols that work at a lot of places, but I don’t hear a lot of people talking about investing in the world that way. I think we really need to take a moment to evaluate our priorities as a global community to be able to make sure that globally we’re strong.

Olivia Allen 22:59
Dr Dennis Chopera, Senior Programme Manager at the leadership for African research networks, and Dr Sophie Gilbert, Programme Lead at the Geneva Science and Diplomacy Anticipator were part of the group that explored what an equitable Life Sciences ecosystem might look like in 25 years time. My colleague Alexandra Canet discussed their perspectives on the future in this space.

Sophie Gilbert 23:18
Yeah, it’s a really interesting question and forces you to think quite creatively about what you imagine a desirable future to look like for yourself. I mean, genomics is such a foundational resource and discipline in the life sciences. What I would hope that would happen in 2050 I would be thrilled if we were starting to generate new medicines and new strategies for health in humans, but also beyond humans in other animals and entities too. Also be able to, if I can expand that analogy to the planet, and hope that we were able to employ the insights we get from genomics to make sure that both the planet and humans who live on it are healthy.

Dennis Chopera 24:14
I completely agree with Sophie. For me, in an ideal world in 2050, I think genomics will be such a powerful, accessible tool for health equity as well as social transformation. I think it will enable precision medicine, including early detection of diseases as well as, you know, personalized therapies that I’d select, but specifically to, you know, accommodate the genetic diversity of different populations. And in this case, I mean not only those populations that are currently well represented in the current data sets, but global populations, including global south and yes, in my line of work, this means harnessing genomics for context specific innovations. This includes solving health challenges that are relevant, in my case, to African populations and developing local solutions and informing equitable policy. For me also, I think for the communities that I serve, it means empowering them. So ultimately, genomics should not be just about technological advancement, but it should be about justice, dignity as well as shared prosperity for all.

Alexandra Canet 25:39
You were both part of a group within the workshop that was looking at this big statement that is Equitable Life Sciences in 2050. So what are the main challenges to achieve this?

Dennis Chopera  25:50
Yes, so for me, I think there are several challenges. So achieving equity by 2050 is quite an ambitious goal, but one that is also essential. What I think is the most pressing challenge really stems from the deep rooted, structural inequities that we have some speaking from the perspective of someone based in the Global South, and I mean inequities such as limited research and limited access to research infrastructure. We’ve also got skewed funding access, as well as under representation in global scientific leadership of the global south. So I think these are all factors that are compounded by the legacies of extractive research practices, where really I mean historically, knowledge flow has been unidirectional and benefiting the Global North.

Alexandra Canet 26:49
This leads us beautifully into my next question, that is, what can institutes, research institutes and universities do? What part can we play in helping achieve this?

Dennis Chopera 26:59
So achieving genomics equity in 2050 requires a lot of change, and universities and academic institutes are quite critical vessels of this change, in the sense that they shape future generations of scientists, as well as influence research culture and in order to foster equity, they must actually be intentional about building inclusive environments that recognise and nature diverse forms of excellence. This includes investing in robust mentorship collaborative research opportunities across different regions of the world, as well as leadership pathways for early career researchers, especially those that are coming from underrepresented regions. I think for early career scientists, it is more than just skills, but it is also about them enhancing their visibility, for their voices to be heard, as well as oppositioning themselves, not just as participants, but also as co- creators in the decision making processes.

Alexandra Canet 28:07
We’ve been touching on universities and you’ve mentioned very briefly policymakers there, Sophie. What do you think the role of policymakers and funders should be in all of this?

Sophie Gilbert 28:19
Yeah, it’s quite interesting actually. Policymaking around the world is seen as quite different. I think in the sciences, we’re seeing much more of a politicisation and maybe even weaponization of science. So countries are seeing science and its scientific resources as essential national strategic assets. And I think this is going up. And I think in Europe, we definitely have this sort of focus on regulation as being the way in which we can ensure equity and make sure that new technologies and new information is used for the global good. But however, I think other regions of the world would see things quite differently. You might want to really create policy to catalyse innovation, because different regional contexts will have very different priorities in that way. And I think this is where policymaking has a massive role, but needs to be both sort of top down in terms of multilateral agreements, so that science is not a national activity. It’s not held nationally either the results or lots of it isn’t and so you need that international cooperation at the policy making level, but absolutely, essentially, you need the bottom up policy making too because you need to be able to set regional priorities and to advance the needs of local populations and what are their concerns and also their priorities. And I think the mixture of the two is going to be essential for achieving equity.

Dennis Chopera 30:06
Yes, as Sophie rightly pointed out, funders have a critical role to play. They must firstly rebalance power by supporting equitable partnerships and simplifying access to funding. And I think they should also invest in long term institutional strengthening in low to middle income countries, as well as simplify funding decision making, which should be shaped by contextual understanding, and should also be co-designed with local stakeholders to ensure relevance as well as sustainability.

Sophie Gilbert 30:41
One example that really springs to mind, which I think is quite illustrative of this question of power and about funding, is that one of the restrictions to sending money outside of the UK from a charitable organisation is on the governance and legal compliance side of funding, which sounds super boring. But it is actually really crucial, because if you want to send money to an institution, say in Southeast Asia, to form the basis of a programme that the university itself is shaping, conducting, holding all the governance for that. The university must be subjected to an audit from the UK, and that university then is subject still to this power dynamic of the UK, being able to assess that University’s operations in a very fully transparent manner. How do we say, no this is not morally acceptable for a higher income country to hold these threads of power at the top and how do we actually give. And I think this is a very difficult question, because it’s going to require really a whole new look at regulatory frameworks, legal frameworks, and these are nationally set. And being able to address power, and to be able to talk about power in this way, I think it would be a really big step forward in the conversation.

Dennis Chopera 32:16
I do agree with you, Sophie, the organisation and the programme that I work with is actually a recipient of quite a lot of funding from the UK. And what you’ve just said really resonates very well with what we go through on a daily basis. Once again, I’m speaking as someone based on the African continent, and we do receive funding from a diverse pool of funders. And we do have funders that have funding models based on merits and potential of the researchers that they are funding, and these will issue out calls and give out funding on a merit basis. But then we also have another group of funders who actually do not have cause. They have, you know, they will identify who they want to find, and in most cases, they do identify people who are researchers, rather, who are already running programmes that were set up by funding from other global funders, and they know very well that the projects have taken off and they are doing so well, and in some cases, they even ask the researchers to rebrand their programmes to align with their own interests. And I think in a nutshell, for us, this really points to alternative drivers of their funding models, which is not really to promote equity. My thinking is that we must first decolonize research, and this means dismantling extractive practices, promoting inclusive governance as well as embedding equity. There’s a core principle in all partnerships that we have, we must also view investment in capacity building not as a charity, but as a strategic imperative for global excellence. And in this, one of the things that we should do is adopt models funding models that you know, promote equity.For example, one of the models that we use in the work that we do here is called hub and spoke model, where the more resourced institutions partner with the less resourced institutions and there is a lot of technology transfer and capacity building at the weaker institutions.Weaker in the sense that they are less resourced. And we’ve seen this to work very well over the past few years, and it has really pushed us quite a long way in achieving equity in research.

Sophie Gilbert 34:55
One of the things I think we can also do now is to anticipate and try to understand what the future of research looks like in this area, in the life sciences, because lots of organisations do have an overview of what’s hot in the field right now, especially the researchers themselves. But that information is not taken out of the academic sphere, and so without it being understood by policymakers, there’s no way for people to be able to coordinate and actually put in place the systems that would enable a more equitable structure for these advances to actually take place in anticipation of them actually becoming concrete realities.

Dennis Chopera 35:45
I just would like to agree with Sophie, and I can give an example. One of the things that we actually require in our programmes is what we call policy engagement and impact storytelling. And this really enables the researchers themselves to actually interact and make themselves visible to policymakers and also articulate the impact of the work that they’re doing in terms of how it impacts policy, and also in terms of driving policies that affect the work that they’re doing.

Olivia Allen 36:26
To close today’s episode, we invited Professor Matt Hurles and Dr Tariq Khokhar, who we heard at the beginning, to engage in a small horizon scanning exercise for their respective institutions. We asked them to paint a picture for you, our listeners, of what their organisations will look like in 25 years time.

Matt Hurles 36:41
I can see us being even more international than we are currently and so just in terms of the diversity of staff that you would see walking in the door. I think currently, our research is very international. So I think last year, we brought in bio samples from well over 100 different countries, not necessarily all humans. It could be species that we’re sequencing, or pathogens. So 80% of our papers are international collaborations. I would certainly see a much higher proportion of those international collaborations would be with lower income countries than they are today. But I think we’d probably see a similar number that were international collaborations, all told. I’d like to think there was more movement of people. I’d like to think of us bringing people into Sanger training them and then them going back to their home countries. We currently do quite a lot of in-country training as well. I think some of that training is done through online platforms, which are very kind of broad, and I think those will increase over time. I think the nature of some scientific training is that it currently really benefits from being side by side with the trainer and so I can well imagine that that’s something that will increase, you know, the amount of in-country training that we do. I certainly think the way in which research ideas are generated will become more equitable. But I think that’s not just about the kind of high income, low income country, that’s also about populations and patient groups and things that they care about.

Tariq Khokhar 38:09
By 2050 I’d want someone to look at welcome and say that’s not just an institution that adapted to the future. It helped them build it, and there’s no point in philanthropy screening organisations that respond to what’s happening. We need to be able to be ahead of those things. So what could that look like? Definitely on the talent side, I think you’d want a Wellcome for us that’s far more permeable with ideas, talent and probably power flowing in from a much broader set of communities than it does today. I think it’s already reasonable to say that the groups that have influence or access to Wellcome decision making and leadership are probably disproportionately even the UK or richer countries. I’d love to see that shift in a sort of substantial way. And I think on similar lines like, you know, creativity and what we do will not just come from funding bolder, riskier, innovative sort of science, but also, as we said, evolving how we support people, how we support institutions. So as we talked about, like, funding different types of institutions, funding different types of careers, understanding that great ideas can come kind of from anywhere, from different combinations of teams, disciplines, etc. Wellcome can’t just be the place that funds academic science. You know our vision is a healthier future for everyone, and funding academic research is not the only route you can take to get there so I think we fully sort of embrace that.

Olivia Allen  39:40
Thank you for listening to the fifth episode of the Genomics Futures podcast. If you haven’t listened to the first four episodes yet, you’ll find them wherever you get your podcasts under Genomics Futures. The next podcast will be the final one in the series, in which we’ll look at the workshop that had a special focus on health and disease. The episode explores the future of personalized medicine, gives a perspective from pharma experts and reflects on the possibilities of genomic sequencing for all. If you want to get in contact, please do. You might agree, disagree, or have your own thoughts about the topics and themes discussed in these conversations. We’d love to hear them. You can get in touch with us at genomicsfutures@sanger.ac.uk.