OmicsXchange episode 2: seventh aniversary of GA4GH — an interview with Peter Goodhand and Ewan Birney

28 Jan 2020

28 January 2020 marks the seventh anniversary of GA4GH — the international standards organisation for genomics. On episode 2 of the OmicsXchange, we’re here to speak with Peter Goodhand, CEO of GA4GH, and Ewan Birney, Chair of the GA4GH Executive Committee, as we reflect on GA4GH’s role — past, present, and future — in advancing genomic and health-related data sharing.

January 28, 2020 marks the 7th anniversary of GA4GH, the international standards organization for genomics. We’re here to speak with Peter Goodhand, CEO of GA4GH, and Ewan Birney, Chair of the GA4GH Executive Committee, as we reflect on GA4GH’s role – past, present, and future – in advancing genomic and health-related data sharing.

Angela Page: Welcome to the OmicsXchange. I’m Angela Page. Today marks the 7th anniversary of the Global Alliance for Genomics and Health – or GA4GH – the international standards organization for genomics. We’re here to speak with Peter Goodhand, CEO of GA4GH, and Ewan Birney, Chair of the GA4GH Executive Committee, as we reflect on GA4GH’s role – past, present, and future – in advancing genomic and health-related data sharing. Before we get started, a little more on our guests: Peter has been with us since GA4GH was founded in 2013, serving first as the Executive Director and now as CEO. He was involved in the initial planning for the organization, before we called ourselves the Global Alliance. Ewan joined as our third Executive Chair in 2016 and was instrumental in launching GA4GH Connect—a new strategic phase for the organization focused on connecting the needs of the broader community with the standards that GA4GH develops. Peter is based at the Ontario Institute for Cancer Research, and Ewan serves as co-Director of the European Bioinformatic s Institute, or EMBL-EBI. So welcome, Peter and Ewan.

Ewan Birney: Thanks for having me.

Peter Goodhand: Great to be here.

Angela Page: GA4GH had its start in 2013 when 50 scientific leaders from eight countries came to New York to discuss the shifting field of genomics. Peter, what was going on in the field to initiate that meeting?

Peter Goodhand: So it was a time of rapid transitioning for probably a couple of years before it people had realized that the drop in sequencing costs, the massive reduction in the cost of genomic sequencing, would take us from a few thousand whole genomes in a dozen institutions to tens of thousands, if not hundreds of thousands of genomes in hundreds of institutions. And so if we could organize if we could collectively organize the community and get ahead of that curve, the chance to create create real value, and if not, we could create a lot of noise and confusion, and slow down the progress from this massive opportunity that was created by lower sequencing costs. People realized we couldn’t keep operating the way we had. We had to modernize genomic research.

Angela Page: And, Ewan, how did you first get involved with GA4GH?

Ewan Birney: So I’ve been involved with big genomics projects for a very, very long time. Iconically, the human genome, and also the Institute, which I’m director of, EMBL-EBI, we had a big stake in making genetics work for very broad set of people. I think we came at it from that – a tradition of international genetics projects and making it work, realizing that there was going to be a shift to more medically-based sequencing. But around 2015, we went through an internal process at EBI at looking at how our science was going to be used inside practicing healthcare, and something that we realized in that analysis was that we were either going to have to create a competitor to GA4GH, or we were going to have to make GA4GH very, very practically work for a large number of groups. We chose the latter – And so many of us across EBI engaged – myself as Chair – but many other groups and if you look at EBI’s engagement with GA4GH, it went through a phase change around 2015, 2016.

Angela Page: So EBI was experiencing this phase change, and then you came on as Chair of GA4GH in 2016 and initiated a phase change here by launching GA4GH Connect. Why was that pivot needed?

Ewan Birney: So there was lots of good work and interesting work happening inside of GA4GH at the time. But I noticed and others that there wasn’t a connection to the incredibly practical questions that many of the people doing genomics in research, and in health care delivery, were facing.

Angela Page: And so that’s what led to the concept of Driver Projects, right? So these are real world projects that both contribute to standards development and help pilot released standards.

Ewan Birney: And of course, we GA4GH get the concrete connection to real world problems, but also those engineers. We have many people who contribute who are not one of the members of these 23 different driver projects. But it has absolutely I think, helped the way GA4GH thinks about standards development.

Angela Page: Peter, you’ve been with GA4GH since it was founded. I’m curious how the organization has changed and evolved over the years?

Peter Goodhand: Yeah, I mean, in the beginning, it was very small. It was a handful of volunteer leaders. It was one or two staff members. It was a coalition of the willing, it was people who really wanted to make a difference, and believe they could do better by working together. The real change over the last several years is we’ve scaled, we’ve got many, many more organizations, we’ve got broader cross section, you’ve got industry strongly represented. We’re more broad, widespread, we’re bigger. But we’ve also particularly since Ewan joined us become much more disciplined, much more structured on the product that we’re actually creating, and we’ve made it a lot more effective for the benefit of community.

Angela Page: And as GA4GH has grown, a key element has been a very strong global community contributing to standards development. How will GA4GH continue to strengthen the global aspect in the future?

Peter Goodhand: We had global there in the name from the beginning. And we have we have worked hard, but it’s been a real effort to make this a truly a global enterprise. And that’s because the the strong contributions in the beginning came from a relatively few countries that have been doing genomics for a long time and doing it at scale. And so there was a natural tendency for those people who’ve been involved in things like the Human Genome Project to continue that collaboration. But we were very committed and very deliberate, to go out and engage more of the global community. And the best way we’ve been able to do that is through growing the breadth and diversity of the driver projects. So the recent edition of the GEM driver project from Japan or H3Africa as a driver project, has given us a real reach into new continents. And I think in Europe, in partnership with ELIXIR, we’ve got a much stronger breadth of participation in Europe, whereas in the beginning, it was just one or two countries. So it’s been a real effort, but I think it’s giving people the opportunity, and then demonstrating value from the participation. We still have to struggle with time zones and travel. But we are, we’re well forward in making it more global than it’s ever been in its existence.

Angela Page: You mentioned the word “Global” was in the name since the beginning, why is it important for a genomic standards organization to be global?

Ewan Birney: There’s two reasons why it’s important. One its what science is about. And in science, we share things globally. We don’t that’s a key fact. It’s something that’s been part of the scientific tradition for a very, very long time, and that would be something bizarre, frankly, to focus on a smaller number of countries or places. The second thing is the opportunity. Ultimately, in every country, probably nearly everyone will be sequenced over this century. It’s a remarkable thing to say. It’s a remarkable thing to imagine. We, as a global scientific community, want to be able to tap into that global opportunity, with benefits flowing back globally to everybody on the planet. It’s a bit cheesy, but it’s nevertheless true.

Angela Page: So in 2017 you and Peter co-authored a paper where you estimated that 60 million clinical genomes would be sequenced by 2025. And we’re almost halfway there, so is your prediction holding steady or would you change it at this point?

Ewan Birney: So it’s a good question. A number of people have asked me whether we should revise that prediction. I think we use the phrase some generic essay, which is a little bit of wiggle room for this, but still, it’s when we when we kind of do a rough benchmarking. I think it’s fair to say we’re we were slightly over optimistic at the time. But my experience of these exponential curves Is that why is very, very hard to judge how it’s going to play out in the future because it’s not the case that these things grow linearly in any shape or form. A good example is the US for Example looking at increased the use of genomics as part of Medicaid in, in the legislation. So I am sticking to the prediction. Because my experience is that it is as easy to underestimate as it is overestimate in these exponential curves. And I suspect will be closer to that number than the benchmarking perhaps suggests.

Peter Goodhand: Yeah, and I would agree, Angela, I think, I certainly know that in the clinical domain with we’re still seeing different uses for large panels for exomes. And for whole genomes, like people are getting much better at understanding which use when and getting the clinical evidence of utility and value. So I think the prediction about the number of genomes is, I think is well within scope. Maybe the way they’re delivered, maybe not exactly the way we could have predicted back in 2017.

Angela Page: So switching gears a bit, I know for a lot of people—at least for me when I first heard about standards—it’s not the most exciting thing in the world. So I’m curious what keeps you both motivated?

Ewan Birney: There is an aspect of standards, which sounds very dry and very uninteresting. It sounds like you’re in a room for a long time with bad lighting and having very detailed discussions. Some of it is like that. But what really motivates me that moment is when you see things click, and there are people where things just work. And then they just get on with their life. And they do that. Every time you pick up your phone, your your mobile phone, and you see your the weather report, or something pings through on your on your phone, you think is, you know, I’ll follow up on that new story. There’s a bunch of standards that have worked, you’re happy you don’t spend a lot of time worrying about it. You don’t say to yourself who set up that standard. They work for you. And in a similar way we want to enable in particular clinicians and researchers to be able to have things that just work, things click together things appear in the right place. And in particular, that’s not just about the kind of protocol to get information around, but it’s also about the representation of data. So you don’t make mistakes. When you have your standards written down correctly, it becomes much harder to make certain classes of mistakes. And that means less bad information will get passed on. better results will happen across health care.

Peter Goodhand: It’s incumbent on us to work with the genomic specialists to work with those established standards organizations that industry and health care rely on and they have to be accredited. So it’s not just the precision and the avoidance of mistakes. These are very necessary elements of operating in, in the industry-setting and in the healthcare-setting. And so the more we can add our specialist knowledge to those established standards, the better for everybody.

Angela Page: As we look towards the future—what is the ideal state of genomic data sharing that you’d like to see GA4GH enable?

Ewan Birney: I think the ideal state is when any researcher across the world has a good idea, good research idea, can easily execute it. That means the analysis, that means the ability to go through what will be necessary checks for doing that research touching different data sets in different places. And that means all the technical components to make that smooth, so that the science isn’t limited by technology. And the science isn’t limited by process. We have to accept the fact that the science will be limited by ethics, legislation and other aspects that are wrapped up with that. But I’ve not met any country or system that doesn’t think that doing research for the benefit of health for the benefit of all humankind is not a good idea. So I don’t think that that although Although ethics and and of course the legalities of what can do trumps research, and I think one has to go back, often to the underpinning aspect that research is a common good for everyone. So I just don’t want the technical aspects to be barriers.

Peter Goodhand: The other paradigm shift that I’m seeing and it is beyond genomic is the move to a learning health system. Much more of a virtuous cycle of exchange of knowledge between research and healthcare and healthcare and research. And as we, as we move into that way of looking at it, it’s no longer about the research finding, taking 10 or 15 years to move into healthcare practice. It’s actually about how from every delivery of healthcare, every episode of healthcare, can we learn and improve the health care of the next individual.

Angela Page: Great – thank you so much Peter and Ewan for joining us here on the OmicsXchange on the 7th Anniversary of GA4GH.

Thank you for listening to the OmicsXchange—a podcast of the Global Alliance for Genomics and Health. The OmicsXchange podcast is produced by Stephanie Li and Caity Forghee, with music created by Rishi Nag. GA4GH is the international standards org for genomics, aimed at accelerating human health through data sharing. I’m Angela Page and this is the OmicsXchange.