Openness, interdisciplinarity and the way forward
The way researchers make discoveries is evolving, and a renewed focus on openness, collaboration and interdisciplinarity is needed to address domestic and international research priorities. Open science principles are central to our vision, and are omnipresent throughout our strategic plan, NSERC 2030: Discovery, innovation, inclusion.
To mark International Open Access Week, we recently caught up with Dr. Timothée Poisot, associate professor with the Department of Biological Sciences at Université de Montréal, to discuss how openness has played a key role in shaping his interdisciplinary research program, and how the current generation of trainees are helping drive the open movement forward.
Can you describe your research program?
I consider myself a computational ecologist. While pursuing my master’s degree in parasitology and epidemiology, I started getting more and more interested in interaction between things in ecology. That is what brought me to Université de Montréal where my lab focuses on developing new quantitative tools for the study of ecosystems.
We look for opportunities to solve broad questions in biodiversity sciences, ecosystem ecology or ecology of pathogens, with tools from other fields, like machine learning, statistics, AI, applied mathematics and supercomputing. The goal is to deliver predictions that will inform policy decisions or that can be used within a decision-making framework.
The Poisot Lab is, first and foremost, an open science lab. Why is openness so important to you and your research?
We try to work in a way that is open by default, which is to say that unless we can find strong arguments to not be entirely transparent about the project that we're working on, everything will be open. For example, you'll see drafts of manuscripts in progress on our GitHub page. You can track the way we write our manuscripts in real time, the development of software packages that we do, or the development of specific pipelines for analysis.
As a publicly funded lab, we want information to be publicly available, and to have more eyes vetting our research. Others working on similar projects can start interacting with what we produce in a meaningful way because they get access to it.
Openness also forces trainees to have very high standards of professionalism, because anyone can go and look at the way they work. This might sound a little terrifying at first; however, the trainees usually get very comfortable with it, and it’s been very empowering for them.
Your research intersects with many other disciplines with different uptakes of openness. Have you found some fields to be more hesitant?
Colleagues in mathematics and physics have been using preprint servers for 25 years, and it’s mostly the norm in fields like machine learning and AI to have code that is open and to work on shared datasets.
Ecology is a bit different and a little more hesitant. Specifically, because the data is logistically and financially difficult to collect. There's a different approach to data ownership that requires more caution in the way we interact with colleagues in our own field. What we try to put forward are the benefits that working in a way that is open will bring to you. We don't necessarily need to see your data, but if we see similar data, we're going to build something that you could use.
Was there an event or a supervisor that led you to become a fervent advocate for open science?
Not really. I'm a first generation academic, so I had no pre-conception of the way things worked. Open access (OA) was a big part of the culture of the lab where I did my PhD. In fact, in year one of my master’s degree, we were some of the first students in France to get a PLOS One t-shirt. I still have my 15-year-old t-shirt. It's full of holes, but it’s like a V1 limited collector’s item!
There was this idea of OA publishing, and I was interested in the open-source movement, therefore I built some of these elements into my practice. It aligned with my personal values as an individual, and I wanted that to be reflected in my work.
How can the research community further promote openness to instill open science as a standard practice?
Baby steps! Christie Bahlai, a Canadian researcher at Kent State University, wrote a great blog post called Baby steps for the Open-Curious. It was about how we need to go very slowly. Open science can rapidly conflict with scientists’ values, or the fears that another scientist could scoop their data. There's this notion that we are competing against one another, and open science counters that. I think we, as open science advocates, need to do a better job of convincing people that it's not that different from the way we are working.
The point is not to bring everyone to doing all open science all the time. You can open things up at different times and there are things that I don't necessarily want to see open. Like people who are working with dual use research concerns. It’s like something you find in a Pandora's box that needs to remain closed.
We also recognize that non-academic stakeholders might have different financial incentives or different constraints regarding intellectual property, data ownership or confidential information. If they say they don't want something to be open, that's fine. Another example is the work that we're doing on beta coronaviruses. We've been careful about possible misinterpretation, so that remains closed until it is published.
Ultimately, open science needs to have some flexibility. Our opinion is that there’s room for compromise and respecting the boundaries of others. We need that.
What do you see as the step or hurdle that we must overcome to push the open movement forward?
The consequence of this movement having been around for 20 years is that students who are starting grad school now, that's all they've ever known in their academic life. As opposed to people who still see it as new, because they have been running labs for longer than that.
For the current generation of grad students, OA is not a disruption. It’s just a status quo, and things should be open. There are way more students than principal investigators in any lab. When they arrive with values about sharing, openness and giving back to the scientific community at large, that shapes the direction of the field. It's very grassroots, where trainees are bringing a lot of culture change.
I don't think that a critical mass of open data or OA journals is going to change everything. It's really changing the values and changing what we consider to be impactful research or fundable research that is going to have more impact.
Where do you hope to go next with your research? How can open science help you in that next phase?
More interdisciplinarity. When I think about a project for grad students and postdocs, I might not be looking for ecologists or biologists to hire, but rather someone from a slightly, or completely, different field, to address challenges that we, as ecologists, have not been trained to tackle, or simply to try different things.
Open science has been a great mechanism to hire students and find collaborators since they engage with our work before engaging with us. Potential students and trainees from all fields can see what we're thinking about. We now receive many spontaneous applications from applicants who’ve seen what we do in GitHub and have thought of specific projects that might evolve into something that could be funded for a postdoc or PhD.
I could write five pages about the way our research program is going to progress, but if students are not fully on board, it's not going to evolve in this direction. It will evolve according to their interests. I mention some boundaries, and they can identify concrete areas where they can contribute. This is very interesting, and it would not have been possible with an entirely closed research model.
What was the significance of NSERC funding to your research program?
My Discovery Grant was the first grant that I received as a principal investigator. It was instrumental in bringing the first grad students on board, and it shaped the direction of the lab for the first 3-4 years before I started exploring different areas. It's important to underline how unrestrictive NSERC is in terms of research directions. I’m also part of two CREATE programs in biodiversity, which help provide adequate support and training to students, while also creating communities of practice, even outside of the lab.
Lastly, how does your research intersect with “Open for Climate Justice,” the theme of this year’s International Open Access Week?
We do some long-term forecasting linked to global climate change and land use change, and we also do work in the context of viral ecology and viral emergence, but it’s very difficult to identify precisely where our work intersects with climate change, because climate change intersects with everything.
We had a discussion in class where I explained that you can train a new model on your data, and one of the students said, “but in five years, depending on the climate change scenario, the model won’t be valid anymore.” That is a fair point. It’s concerning, and it is permeating everything we do as ecologists and biodiversity scientists. It’s this looming existential threat that is here all the time.
This interview has been condensed and edited for length and clarity.
Indigenous U of G prof weaves knowledge systems in unique research on Eastern wolves
The first thing University of Guelph professor Dr. Jesse Popp did when considering how to research the Eastern wolf population in Northern Ontario was to reflect on what years of relationship building with local Indigenous communities had provided.