Understanding Open Science

Colette E. Bean, MA, APS chief publishing officer, spoke with Tracey Weissgerber, PhD, a member of the APS Publications Committee and group leader at the QUEST Center for Responsible Research at the Berlin Institute of Health at Charité. Weissgerber is a physiologist who now studies research itself. She looks at ways to make research more transparent, rigorous, reproducible, useful and trustworthy to science and society.

Bean: Could you explain your view on what open science is and give a few examples?

Weissgerber: At its most basic level, open science is about making our materials and our products of the scientific work that we do more open, more available and accessible to others, and making our scientific process more transparent. That might include sharing the data that we generated on a data repository so that others can reuse it for different studies. It might include sharing the code that you use to analyze the data. It can also include making software, or a tool that you have developed, openly available to others, or making educational resources that you’ve created open so that others can reuse them.

One of the things that’s really critical to open science is that we diversify our understanding about what a research output is. Right now, our system is very focused on publications, which are primarily about the results. But when we think about it, a lot of the work really goes into creating these many other pieces—the data, the code, developing the methods and the procedures. And so, allowing [our colleagues] to reuse our methods, our data, and to cite those materials so that the researchers who’ve created those materials can get credit for their work, is really important.

The other piece of open science is about scientific rigor and reproducibility. Just making something open doesn’t mean that it’s high quality. We also want to know about the experimental design. We want to know if the authors are using procedures like blinding or randomization or pre-registration to help us evaluate the risk of bias and the quality of the way that the study was done. This information is really important because it tells us how useful the results might be, how much confidence we should have in those results, what kind of follow-up work might be needed or nuance in the interpretation. 

Bean: Can you talk about the importance of open science as it relates specifically to biomedical research? 

Weissgerber: One of the things that’s especially important about biomedical research is that we ultimately want therapies that work for patients. We want things that translate and that solve problems in the world. So it’s really important to go beyond thinking about “What do I need to do as a scientist to build my career?” And also to think about “What is the end goal of this? Who is my research trying to help?” … Making our materials open, making them high quality, making them rigorous and trustworthy and useful to others can help us accelerate that process of discovery. 

But in order to do that, we have to change the way that we assess scientists. We need to go beyond rewarding people for their grants and their publications, and we need to start also rewarding them for those other types of research outputs they produce, [such as data sets, software and educational tools]. 

Right now, there’s one way to be a successful scientist, and it’s about grants and papers, which focus on results. And when you’re focusing just on the results, you incentivize people to keep their methods and their data private because they want to have those things to generate more new results as a competitive advantage. If we were to open up what we think of as a research output and start giving credit to people who are good in these other areas, we would provide more paths for being successful scientists. That’s really fundamental for allowing science to advance quickly and efficiently.

Bean: What do you think are the most impactful open science practices and recommendations?

Weissgerber: Right now in open science there’s a lot of emphasis on open access, which is publications, [and on] open data. What we’re not hearing much about is open methods. And I think this is a huge oversight for several reasons. First of all, reproducibility starts with methods. So if we don’t know what you did, we can’t reproduce it and we can’t implement it in another setting. The second reason is that if you care about open data and you want to be able to reuse that data, you have to understand how it was generated to use it responsibly. And then the third reason is that of all of the things that science produces, methods are perhaps one of the most useful. It’s more likely that someone might be able to reuse your methods or adapt them for another context than they would be able to reuse your data.

I would really encourage scientists to look at how to use protocol repositories, how to share their reusable step-by-step protocols and publish their study design protocols so that others can understand and use these things. I’m currently organizing a group called PRO-MaP, which stands for Promoting Reusable and Open Methods and Protocols. We have just released draft recommendations for actions that four stakeholder groups can take. 

Bean: What are some of the challenges of open science?

Weissgerber: The first big challenge is that right now it’s not adequately rewarded or incentivized. So if we think about making our data [and protocols] available … each of those steps take time. We’re still not adequately rewarding people who take the extra time to do these practices. The second challenge we have is a knowledge gap. A lot of people simply aren’t aware of practices [or tools] they could be using. There’s a lot of training to be done to get people to understand, “How do I use these tools and how does that benefit my research group?”

The last thing is resources. Some of these skills also take resources to implement. You may need infrastructure [and] personnel, and there are costs associated with that. The challenge for scientists is if they have [only] so much funding on a grant, they always want to put it into the science. So, thinking about a change in the way that we fund science is really important to make sure that we are leaving adequate resources to support investigators in doing reproducible, rigorous and open research work.

Bean: Do you have any last thoughts?

Weissgerber: Some of this is also about normalizing a culture of transparency and openness. And part of that is understanding that science is messy and it’s very complicated and we all make mistakes. … My recommendation for people would be to try new things. We have to understand that science is evolving and what was considered great scientific practice 20 years ago is different from today because we have new capabilities and new tools. Trying out something new doesn’t mean that you’ve been doing it wrong for the last 20 years. It just means that science is changing, and we all want to change with science so that we’re doing the best research that we can do. 

This article was originally published in the September 2023 issue of The Physiologist Magazine.