Gold Open Access—the model where authors (or their funders) pay article processing charges to make work freely available—has grown steadily over the past decade. But if you look closely, the growth is slowing. And what’s taking its place isn’t the altruistic, community-powered model many hoped for. Instead, Hybrid Open Access is filling the gap: a model where paywalled journals charge extra to make select articles open.

Diamond Open Access—where publishing is free for both authors and readers, supported by institutions rather than APCs—has been making headlines. But here’s the uncomfortable truth: we still don’t have enough data to know whether Diamond OA is actually growing or simply being talked about more.

A System Stuck in the Middle

Digging into Dimensions data layered on top of OpenAlex, a clear pattern emerges. Gold OA’s momentum is slowing. But instead of researchers turning to Green OA (self-archiving in repositories) or Diamond OA, many are choosing Hybrid OA instead.

Why?

Because researchers chase visibility, reputation, and prestige. That often means publishing in high-impact journals—many of which are owned by legacy publishers who now offer hybrid models. These options give authors a way to comply with funder mandates without sacrificing perceived academic clout.

Institutions and libraries, meanwhile, are under pressure to show open access progress. Hybrid OA, while expensive, is a politically safe way to do that. It’s the administrative equivalent of checking a box—even if it means paying twice.

Transformative agreements like Read and Publish deals have only accelerated this trend, redirecting subscription budgets to cover OA fees—effectively normalizing hybrid publishing in many disciplines.

A New Vision Emerges

But while the hybrid tide rises, a quiet revolution is underway.

In 2022, a coalition of organizations—Science Europe, cOAlition S, OPERAS, and ANR—launched a bold Action Plan to support Diamond OA. Their goal: to build a truly equitable, community-driven publishing ecosystem, where knowledge is a public good and the costs are shouldered collectively—not by individual researchers.

This vision took shape through the DIAMAS project and culminated in the creation of the Diamond OA Standard (DOAS). Think of DOAS as a blueprint: a framework to help Diamond journals measure, improve, and sustain quality.

It’s built on seven pillars:

• Legal ownership, mission, and governance
• Open science practices
• Editorial management and research integrity
• Technical service efficiency
• Visibility and impact
• Equity, diversity, inclusion, and multilingualism
• Continuous improvement

Together, these components aim to professionalize Diamond OA without compromising its values. They send a clear message: if scholarly communication is a public good, then it must be shaped and governed by the scholarly community itself.

The Missing Link: Measurable Growth

Despite this momentum, the numbers tell a more sobering story.

Early data from OpenAlex paints the picture that Diamond OA has plateaued in terms of publication volume. Enthusiasm and infrastructure have grown—but the data doesn’t reflect this.

Why the disconnect? Part of the issue is visibility. At Digital Science, we would love a better way to track Diamond Open Access growth. DOAJ lists 1,369 journals as being “without fees”, but there seem to be many edge cases resulting in a landscape that isn’t black and white.

It is also true that many Diamond journals operate with limited marketing, uncertain technical infrastructure, and fragmented funding. And despite the ideals, many researchers still don’t see them as viable options for career advancement.

What Comes Next?

There are reasons for optimism. The DIAMAS project isn’t just advocating for Diamond OA—it’s building the scaffolding. Its service portal offers templates, best practices, and technical guidance to help journals align with DOAS standards and professionalize their operations. The European Diamond Capacity Hub (EDCH) serves as a coordination center for Diamond OA stakeholders in Europe. Launched alongside the EDCH, the ALMASI Project focuses on understanding non-profit OA publishing in Africa, Latin America, and Europe.

The pieces are falling into place for equitable open access solutions. What’s needed is adoption and quantification. Funders and Institutions should include equitable OA in their promotion criteria. Researchers must see it as a credible home for their work. If anyone knows of faster ways to get clean data for Diamond, please reach out.

The path forward exists. But like any path, it only becomes clear by walking it. If you are working in this space and have this data, we would love to disseminate it through our tools at Digital Science.

The post From Gold to Diamond: Is Equitable Open Access Still a Mirage? appeared first on Digital Science.

AI can also serve as a tool to aid humans in data extraction from many papers. Consider a scenario where AI synthesizes information from hundreds of studies to create a refined dataset. That dataset then feeds back into the system, sparking new research papers.

This cycle—dataset to paper, paper to knowledge extraction, knowledge to new datasets—propels an accelerating loop of discovery. Instead of a linear research pipeline, AI enables a continuous, self-improving knowledge ecosystem.

From data to papers

I looked for interesting datasets on Figshare. The criteria was a) that I knew they would be re-usable as they had been cited several times. And b) the files were relatively small (<100MB) so as not to hit the limits of the common AI tools. 

This one fit the bill:

Rivers, American (2019). American Rivers Dam Removal Database. figshare. Dataset. https://doi.org/10.6084/m9.figshare.5234068.v12

From there I asked Claude 3.7 Sonnet “Based on the attached files, can you create a full length academic paper with an abstract, methods results, discussion and references”. Followed by “Can you convert the whole paper to latex so I can copy and paste it into Overleaf?”

The resulting paper needs a little tweaking in the layout of the results and graphs, but other than that, has done a great job.

Papers to new data/knowledge

A single paper is just the beginning. The real challenge is synthesizing knowledge from the ever-growing volume of research. This is where specialized knowledge extraction tools become crucial. How do we effectively mine this knowledge? This is where ReadCube shines. ReadCube helps researchers manage and discover scholarly literature, but its real power lies in its knowledge extraction capabilities. Imagine ReadCube as a powerful filter, sifting through countless pages to extract the nuggets of wisdom.

Tools like ReadCube can then analyze vast collections of papers, uncovering patterns and relationships that human researchers might miss. This process involves:

• Text and citation mining: AI can analyze papers to identify emerging trends, inconsistencies, or knowledge gaps.
• Automatic synthesis: AI can compare findings across thousands of studies, synthesizing insights into new, high-level conclusions.
• Hypothesis generation: By recognizing correlations between disparate research areas, AI can propose new research directions.
  

The Flywheel Effect: How the Cycle Accelerates

The true magic happens when this extracted knowledge becomes the input for the next iteration. Each cycle follows this pattern:

1. Raw data is processed by AI to generate initial research outputs
2. Knowledge extraction tools mine these outputs for higher-order insights
3. These insights form a new, refined dataset
4. AI processes this refined dataset, generating more precise analyses
5. The cycle continues, with each rotation producing more valuable knowledge

With each turn of this flywheel, the insights become more refined, more interconnected, and more actionable. The initial analyses might focus on direct correlations in the data, while later iterations can explore complex causal relationships, predict future trends, or suggest optimal intervention strategies.

This AI-driven, data-to-knowledge cycle represents a paradigm shift in research. Imagine the possibilities in fields like medicine, climate science, and economics. We’re moving towards a future where AI and human researchers work in synergy, pushing the boundaries of discovery. Rather than replacing researchers, AI acts as a force multiplier, enabling deeper faster knowledge generation.

The post The Perpetual Research Cycle: AI’s Journey Through Data, Papers, and Knowledge appeared first on Digital Science.

Every Muppets fan knows that Christmas is all about being revisited by people you’ve previously encountered. So from 25th December to 5th January we’ll be sharing our 12 Days of DSmas. Check back daily as we share a Speaker Series chat each and every day. Happy Holidays from the Digital Science Thought Leadership Team!

Merry Dr Chris Van Tulleken-mas! We chatted with Chris online about research integrity, impact, openness, and investigative research. Catch his interview here, and don’t forget to watch his Xmas Lectures on BBC for The Royal Institution this year!

Chris on Spotify

Chris on Apple Podcasts

Chris on YouTube

Read more

Professor Dame Athene Donald joined our Speaker Series line-up in February after her Speaker Series Live talk at the Ri in London, UK. Here she is talking about why we need to empower women in STEM from the ground up.

Athene on Spotify

Athene on Apple Podcasts

Athene on YouTube

Read more

It’s time to go to Berlin, Germany and catch Matyas Kovacs at the Falling Walls Summit. Matyas shared his thoughts on communication and innovation to impact all of society with science.

Matyas on Spotify

Matyas on Apple Podcasts

Matyas on YouTube

Read more

As a Nobel laureate and former president of The Royal Society, Professor Venki Ramakrishnan has long played a role in shaping a more innovative, inclusive and impactful research culture, which we chatted about during his live Speaker Series lecture at the Ri. We went to Cambridge, UK to hear his thoughts on curiosity, competition and collaboration.

Venki on Spotify

Venki on Apple Podcasts

Venki on YouTube

Read more

As Chief Publishing Officer at PLOS, Niamh provides business leadership for the entire PLOS portfolio to advance PLOS’s vision and mission. In this episode Niamh talks about the evolving landscape of scientific research and the push towards open science, including her journey from the early days of advocating for public access to research, to tackling current challenges like making science more inclusive and accessible.

Niamh on Spotify

Niamh on Apple Podcasts

Niamh on YouTube

Read more

Building communities is hard, but Alice Meadows has worked hard to make it look effortless. Here she is in Boston, MA, USA, telling us about the power of persistent identifiers.

Alice on Spotify

Alice on Apple Podcasts

Alice on YouTube

Read more

It’s New Year’s Eve, and a time to reflect on the past and make plans for the months ahead. When we visited the Max Planck Institute in Berlin, Germany, we added to the echoes of amazing research conversations resonating around their iconic library when we chatted about the history, philosophy and future of research with Dr Maria Avxentevskaya and Dr Ben Johnson.

Maria and Ben on Spotify

Maria and Ben on Apple Podcasts

Maria and Ben on YouTube

Read more

Happy New Year! We caught up with pro-skater Rodney Mullen at his home in Los Angeles, USA to hear his thoughts on why we need diverse minds to innovate in all walks – and ollies – of life. And, since it’s the new year and you’re probably feeling a little “sleep deprived”, you can also follow this up with his live Speaker Series lecture at the Ri.

Rodney on Spotify

Rodney on Apple Podcasts

Rodney on YouTube

Read more

If you’ve been eating as much cheese as this author, dearest gentle reader, you too will be experiencing a fascinatingly slippery grasp on reality – which brings us to Day 9’s featured speaker. “Is Maths Real?” was the question that Dr Eugenia Cheng posed in her live Speaker Series lecture at the Ri. I caught up with her ahead of her lecture in the iconic Faraday lecture theatre in London, UK to talk about why we need to break down barriers of knowledge in research, and reunite STEM and the humanities for impactful change.

Eugenia on Spotify

Eugenia on Apple Podcasts

Eugenia on YouTube

Read more

Dr Gilbert De Gregorio leads the Frontiers Planet Prize. I caught up with him in Berlin, Germany, to talk about why we need more inclusive competitions to catalyse innovation that impacts our global society.

Gilbert on Spotify

Gilbert on Apple Podcasts

Gilbert on YouTube

Read more

2024 was a wild ride for global politics, and research is not immune to its changes. I caught up with Professor Jenny Reardon in Cambridge, UK, to learn more about how we can work with politics, and not against it, to provide solutions for everyone across the world, and where red tape remains to be overcome.

Jenny on Spotify

Jenny on Apple Podcasts

Jenny on YouTube

Read more

Our final Speaker Series guest of 2024 was Dr Danny Hillis. We visited the Applied Invention offices in Cambridge, MA, USA, where innovator, inventor, and Imagineer Danny shared his thoughts on how we can use novel technology to combat novel challenges in mis- and disinformation and make the most meaningful impact from data.

Danny on Spotify

Danny on Apple Podcasts

Danny on YouTube

Read more

The post The 12 Days of DSmas appeared first on Digital Science.

Joy believes that open infrastructure can tackle both of these challenges and has already gone some way to proving its value in connecting people to underexplored areas of research and underrepresented researchers. Open infrastructures can also help us better analyse and understand the networks forming within and beyond research to drive research, discovery and innovation.

Reflecting on the impact that open research infrastructures have already had on the visibility and the collaborative discovery potential of African research outputs, Joy knows first-hand that these new ways of supporting research are already working – but she encourages us to create an even stronger framework within which everyone’s research contributions can be seen and built upon.

Subscribe now to be notified of each weekly release of the latest TL;DR Short, and catch up with the entire series here. 

If you’d like to suggest future contributors for our series or suggest some topics you’d like us to cover, drop Suze a message on one of our social media channels and use the hashtag #TLDRShorts.

The post TL;DR Shorts: Joy Owango on Research Infrastructure appeared first on Digital Science.

Venki is a strong supporter of open research. Most research activities undertaken across the world are funded by taxpayers or charities. Venki believes that the public and other researchers have a right to access research information generated from this work. However, he also has concerns about the potential consequences of cutting out some of the steps in more traditional publishing routes.

Venki emphasises the need for research outputs to be checked for their integrity and credibility when sharing research openly. This curation step takes time and resources and therefore costs money, so we need to find a balance between sharing research information openly, widely and without financial barriers to accessing it, while also ensuring that that research is robust and of high quality. Who should fund this?

One way the open research agenda has transformed research is through institutions and funders insisting that data sets are made more available by sharing them in open repositories. This is just one example of how culture change in research is happening, and how small steps lead to a much larger research transformation and a paradigm shift in how we do research.

Subscribe now to be notified of each weekly release of the latest TL;DR Short, and catch up with the entire series here. 

If you’d like to suggest future contributors for our series or suggest some topics you’d like us to cover, drop Suze a message on one of our social media channels and use the hashtag #TLDRShorts.

The post TL;DR Shorts: Venki Ramakrishnan on Open Research appeared first on Digital Science.

As part of our ongoing investigation into Research Transformation, we are delighted to present a new report, Research Transformation: Change in the era of AI, open and impact.

Within the report, we sought to understand from our academic research community how research transformation is experienced across different roles and responsibilities. The report, which is a mixture of surveys and interviews across libraries, research offices, leadership and faculty, reflects transformations in the way we collaborate, assess, communicate, and conduct research.

The positions that we hold towards these areas are not the same as those we held a decade or even five years ago. Each of these perspectives represent shifts in the way that we perceive ourselves and the roles that we play in the community. Although there is concern about the impact that AI will have on our community, our ability to adapt and change is reflected strongly across all areas of research, including open access, metrics collaboration and research security. That such a diverse community is able to continually adapt to change reflects well on our ability to respond to future challenges.

Key findings from the report:

• Open research is transforming research, but barriers remain

• Research metrics are evolving to emphasize holistic impact and inclusivity

• AI’s transformative potential is huge, but bureaucracy and skill gaps threaten progress

• Collaboration is booming, but increasing concerns over funding and security

• Security and risk management need a strategic and cultural overhaul

We do these kinds of surveys to understand where the research community is moving and how we can tweak and adapt our approach as a company. We were very grateful to the great minds who helped us out with a deep dive into what has affected their roles and will affect their roles going forward. Metrics, Open Research and AI are very aligned with the tools that we provide for academics, and the strategy we have to make research more inclusive, transparent and trustworthy.

The post Presenting: Research Transformation: Change in the era of AI, open and impact appeared first on Digital Science.

Joy believes that open access is one of the best things that has ever happened for researchers. It has raised the visibility of African research outputs by reducing or removing blockers to the inclusion of the amazing and impactful research being carried out, such as the fees associated with publishing novel research in top-tier journals with broad engagement. Publishers play a key role in driving this change, and overcoming this challenge requires them to fundamentally change how they work with the research community across the world.

Joy reminds us that there are also unexpected consequences that must be overcome as a global research community. An open research culture should allow more people to access existing research information and also contribute to it without financial or other barriers. While preprints and APC waivers exist, there is still a long way to go to ensure that all actors across the research ecosystem are taking advantage of these programmes of research transformation to break down barriers to inclusion in research and foster greater translation of research into real-world impact.

Subscribe now to be notified of each weekly release of the latest TL;DR Short, and catch up with the entire series here. 

If you’d like to suggest future contributors for our series or suggest some topics you’d like us to cover, drop Suze a message on one of our social media channels and use the hashtag #TLDRShorts.

The post TL;DR Shorts: Joy Owango on Open Research appeared first on Digital Science.

Open research is transforming the way research findings are discovered, shared and reproduced. As part of our commitment to the Open Principles and research transformation, we are looking into how open research is transforming roles, approaches, policies and, most importantly, mindsets for everyone across the research landscape. See our inspiring transformational stories so far.

Academia is at a pivotal juncture. It has often been criticized as slow to change, but external pressures from an increasingly complex world are forcing rapid change in the sector. To understand more about how the research world is transforming, what’s influencing change, and how roles are impacted, we reached out to the research community through a global survey and in-depth interviews.

Research Transformation stories so far…

Digital Science Open Principles

Read now

Open Access: Mo Money Mo Problems

Read now

The Initial Transformation

Read now

Academic Survey Report Pre-registration

Sign-up

State of Open Data 2024 – Special Edition

Sign-up

Will 2025 be a turning point for Open Access? – Digital Science

Read now

How has innovation shaped Open Research? What does the future hold – especially with the impact of AI? Here’s Dan Valen speaking about Figshare’s key role, with innovation helping to transform the research landscape.

Digital Science has always understood its role as a community partner – working towards open research together. Here’s some ways in which we have helped to transform research over the last 14 years.

In our first piece, Simon Porter and Mark Hahnel introduce the topic and detail the three areas the campaign will focus on.

The Story of Data Transformation

The Story of Connection

The Story of Research Innovation

• Making data more usable
• Opening up channels & the flow of information
• Transforming data through innovation & AI
• Maintaining trust & integrity

• Seeing both perspectives
• What success looks like for knowledge transfer
• Evolving roles and the role of people in bridging gaps

• Research Transformation White Paper
• How have roles changed:
  • In Academia?
  • In Publishing?
  • In Industry?
• State of AI Report
• How are we using AI in our research workflows?

Research Transformation

The way we interact with information can amplify our ability to make connections, and in doing so transforms how we understand the world. Supercharged by the AI moment that we are in, the steady march of digital transformation in society over the last three decades is primed for rapid evolution. What is true for society, is also doubly so for research. Alongside ground-breaking research and discoveries is the constant invitation to adapt to new knowledge and abilities. Combine the general imperative within the research sector to innovate with the rapidly evolving capabilities of generative AI and it is safe to say that expectations are high. Taking effective advantage of new possibilities as they arise however, requires successful coordination within society and systems. 

There is an art to transformation, and understanding the mechanisms of transformation places us in the best position to take advantage of the opportunities ahead.

In this series, we specifically seek to explore Research Transformation with an eye to adapting what we already know to the present AI moment. Transformation in Research is not just about digital systems, but it is also about people and organisations – crossing boundaries from research to industry, emerging new research sectors, creating new narratives and adapting to the possibilities that change brings.

At Digital Science, we have always sought to be an integral part of research transformation, aiming to provide products that enable the research sector to evolve research practice – from collaboration and discovery through to analytics and administration. Our ability to serve clients from research institutions to funders, publishers, and industry has placed us in a unique position to facilitate change across the sector, not simply within silos, but between them. In this series, we will be drawing on our own experiences of research transformation, as well as inviting perspectives from the broader community. As we proceed we hope to show that Research Transformation isn’t just about careful planning, but requires a sense of playfulness – a willingness to explore new technology, a commitment to a broader vision for better research, as well as an ability to build new bridges between communities.

In the first of three themes, we will cover Research Transformation from the perspective of the data and metadata of research. How do changes to the metadata of research transform our ability to make impact, as well as see the research community through new lenses? How does technology enable these changes to occur? Starting almost from the beginning, we will look at how transitions in publishing practice have enabled the diversity of the research workforce to become visible. We will also trace the evolving story of the structure of a researcher’s papers, from the critical use of identifiers, to adoption of the credit ontology, through to the use of trust markers (including ethics statements and data and code availability, and conflict of interest statements.) The evolving consensus on structured and semi structured nature of research articles changes not only the way we discover, read and trust individual research papers, but also transforms our ability to measure and manage research itself.

Our focus will not only be reflective, but will also look forward to the emerging challenges and opportunities that generative AI offers. We will ask deep questions about how research should make its way into large language models. We also explore the new field of Forensic Scientometrics that has arisen in response to the dramatic increase in bad faith science in part enabled by generative AI, and the new research administration collaborations that this implies – both with research institutions and across publishing. We will aso offer more playful, experimental investigations.  For example, a series on ‘prompt engineering for librarians’ draws on the original pioneering spirit of the 1970’s MEDLARS Analysts to explore the possibilities that tools such as OpenAI can offer. 

Lifting up from the data, we note that a critical part of our experience of research transformation has been the ability to experience and connect with research fromshifting perspectives. In this second theme exploring research transformation, we aim to celebrate the art of making connections, from the personal transformations required  to make the shift from working within research institutions to industry, through to the art of building research platforms that support multiple sectors. We also cover familiar topics from new angles, For instance, how do the FAIR data principles benefit the pharmaceutical industry? How do we build effective research collaborations with emerging research sectors in Africa?

In our third theme, we will explore Research Transformation from the perspective of innovation, and how it has influenced the way research is conducted. Culminating in a  Research Transformation White Paper we will explore how roles have changed in academia, publishing, and industry.  Within this broader context of Research transformation, we ask ‘How are we using AI in our research workflows?’ How do we think we will be using AI in years to come?

Of course, many of us in the Digital Science community have been engaging with different aspects of research transformation over many years. If you are keen to explore our thinking to date, one place that you might like to start is at our Research Transformation collection on Figshare. Here we have collated what we think are some of our most impactful contributions to Research Transformation so far. We are very much looking forward to reflecting on research transformation throughout the year. If you are interested in contributing, or just generally finding out more, why not get in touch?

The post Welcome to… Research Transformation!  appeared first on Digital Science.

Inspired by bug bounty programs in the tech industry, ERROR offers financial rewards to those who identify and report errors in academic research. ERROR has the potential to revolutionize how we approach, among other things, research integrity and open research by incentivizing the thorough scrutiny of published research information and enhancing transparency.

I sat down with two other members of the TL;DR team, VP of Research Integrity Leslie McIntosh and VP of Open Research Mark Hahnel, to shed light on how ERROR can bolster trust and credibility in scientific findings, and explore how this initiative aligns with the principles of open research – and how all these things can drive a culture of collaboration and accountability. We also discussed the impact that ERROR could have on the research community and beyond.

Leslie’s perspective on ERROR

Leslie’s initial thoughts about ERROR were cautious, recognizing its potential to strengthen research integrity but also raising concerns about unintended consequences.

She noted that errors are an inherent part of the scientific process, and over-standardization might risk losing the exploratory nature of discovery. Drawing parallels to the food industry’s pursuit of efficiency leading to uniformity and loss of nutrients, Leslie suggested that aiming for perfection in science could overlook the value of learning from mistakes. She warned that emphasizing error correction too rigidly might diminish the broader mission of science – discovery and understanding.

Leslie: “Errors are part of science and part of the discovery… are we going so deep into science and saying that everything has to be perfect, that we’re losing the greater meaning of what it is to search for truth or discovery [or] understand that there’s learning in the errors that we have?”

Leslie also linked this discussion to open research. While open science encourages interpretation and influence from diverse participants, the public’s misunderstanding of scientific errors could weaponize these mistakes, undermining trust in research. She stressed that errors are an integral, even exciting, part of the scientific method and should be embraced rather than hidden.

Mark’s perspective on ERROR

Mark’s initial thoughts were more optimistic, especially within the context of open research.

Mark: “…one of the benefits of open research is we can move further faster and remove any barriers to building on top of the research that’s gone beforehand. And the most important thing you need is trust, [which] is more important than speed of publication, or how open it is, [or] the cost-effectiveness of the dissemination of that research.”

Mark also shared his excitement about innovation in the way we do research. He was particularly excited about ERROR’s approach to addressing the problem of peer review, as the initiative offers a new way of tackling longstanding issues in academia by bringing in more participants to scrutinize research.

He thought the introduction of financial incentives to encourage error reporting could lead to a more reliable research landscape.

“I think the payment for the work is the most interesting part for me, because when we look at academia and perverse incentives in general, I’m excited that academics who are often not paid for their work are being paid for their work in academic publishing.”

However, Mark’s optimism was not entirely without wariness. He shared Leslie’s caution about the incentives, warning of potential unintended outcomes. Financial rewards might encourage individuals to prioritize finding errors for profit rather than for the advancement of science, raising ethical concerns.

Ethical concerns with incentivization

Leslie expressed reservations about the terminology of “bounty hunters”, which she felt criminalizes those who make honest mistakes in science. She emphasized that errors are often unintentional.

Leslie: “It just makes me cringe… People who make honest errors are not criminals. That is part of science. So I really think that ethically when we are using a term like bounty hunters, it connotes a feeling of criminalization. And I think there are some ethical concerns there with doing that.”

Leslie’s ethical concerns extended to the global research ecosystem, noting that ERROR could disproportionately benefit well-funded researchers from the Global North, leaving under-resourced researchers at a disadvantage. She urged for more inclusive oversight and diversity in the initiative’s leadership to prevent inequities.

She also agreed with Mark about the importance of rewarding researchers for their contributions. Many researchers do unpaid labor in academia, and compensating them for their efforts could be a significant positive change.

Challenges of integrating ERROR with open research

ERROR is a promising initiative, but I wanted to hear about the challenges in integrating a system like this alongside existing open research practices, especially when open research itself is such a broad, global and culturally diverse endeavor.

Both Leslie and Mark emphasized the importance of ensuring that the system includes various research approaches from around the world.

Mark: “I for one think all peer review should be paid and that’s something that is relatively controversial in the conversations I have. What does it mean for financial incentivization in countries where the economics is so disparate?”

Mark extended this concept of inclusion to the application of artificial intelligence (AI), machine learning (ML) and large language models (LLMs) in research, noting that training these technologies requires access to diverse and accurate data. He warned that if certain research communities are excluded, their knowledge may not be reflected in the datasets used to build future AI research tools.

“What about the people who do not have access to this and therefore their content doesn’t get included in the large language models, and doesn’t go on to form new knowledge?”

He also expressed excitement about the potential for ERROR to enhance research integrity in AI and ML development. He highlighted the need for robust and diverse data, emphasizing that machines need both accurate and erroneous data to learn effectively. This approach could ultimately improve the quality of research content, making it more trustworthy for both human and machine use.

Improving research tools and integrity

Given the challenges within research and the current limitations of tools like ERROR, I asked Leslie what she would like to see in the development of these and other research tools, especially within the area of research integrity. She took the opportunity to reflect on the joy of errors and failure in science.

Leslie: “If you go back to Alexander Fleming’s paper on penicillin and read that, it is a story. It is a story of the errors that he had… And those errors were part of or are part of that seminal paper. It’s incredible, so why not celebrate the errors and put those as part of the paper, talk about [how] ‘we tried this, and you know what, the refrigerator went out during this time, and what we learned from the refrigerator going out is that the bug still grew’, or whatever it was.

“You need those errors in order to learn from the errors, meaning you need those captured, so that you can learn what is and what is not contributing to that overall goal and why it isn’t. So we actually need more of the information of how things went wrong.”

I also asked Mark what improvements he would like to see from tools like ERROR from the open research perspective. He emphasized the need for better metadata in research publishing, especially in the context of open data. Drawing parallels to the open-source software world, where detailed documentation helps others build on existing work, he suggested that improving how researchers describe their data could enhance collaboration.

Mark also feels that the development of a tool like ERROR highlights other challenges in the way we are currently publishing research, such as deeper issues with peer review, or incentives for scholarly publishing.

Mark: “…the incentive structure of only publishing novel research in certain journals builds into that idea that you’re not going to publish your null data, because it’s not novel and the incentive structure isn’t there. So as I said, could talk for hours about why I’m excited about it, but I think the ERROR review team have a lot of things to unpack.”

Future of research integrity and open research

What do Leslie and Mark want the research community to take away from this discussion on error reporting and its impact on research integrity and open research?

Leslie wants to shine a light on science communication and its role in helping the public to understand what ERROR represents, and how it fits into the scientific ecosystem.

Leslie: “…one of the ways in which science is being weaponized is to say peer review is dead. You start breaking apart one of the scaffolds of trust that we have within science… So I think that the science communicators here are very important in the narrative of what this is, what it isn’t, and what science is.”

Both Leslie and Mark agreed that while ERROR presents exciting possibilities, scaling the initiative remains a challenge. Mark raised questions about how ERROR could expand beyond its current scope, with only 250 papers reviewed over four years and each successful error detection earning a financial reward. Considering the millions of papers published annually, it is unclear how ERROR can be scaled globally and become a sustainable solution.

Mark: “…my biggest concern about this is, how does it scale? A thousand francs a pop, it’s 250 papers. There [were] two million papers [published] last year. Who’s going to pay for that? How do you make this global? How do you make this all-encompassing?”

Conclusion

It is clear from our discussion that ERROR represents a significant step forward in experimenting to enhance both research integrity and open research through this incentivised bug-hunting system.

Leslie has highlighted how the initiative can act as a robust safeguard, ensuring that research findings are more thoroughly vetted and reliable, but she does remind us that we need to be inclusive in this approach. Mark has also emphasized the potential for a tool like this in making publication processes more efficient – and even finally rewarding researchers for all the additional work that they’re doing – but he does wonder how this can scale up to foster a more transparent and collaborative research environment that aligns perfectly with the ethos of open research as well.

Leslie and Mark’s comments are certainly timely, given that the theme of Digital Science’s 2024 Catalyst Grant program is innovation for research integrity. You can find out more about how different segments of research can and should be contributing to this space by reading our TL;DR article on it here.

We look forward to exploring more innovations and initiatives that are going to shape – or shatter – the future of academia, so if you’d like to suggest a topic we should be discussing, please let us know.

The post The TL;DR on… ERROR appeared first on Digital Science.

Drug discovery has come a long way. It started with natural remedies and saw landmark serendipitous discoveries like penicillin in 1928. The mid-20th century introduced rational drug design, targeting specific biological mechanisms. Advances in genomics, high-throughput screening, and computational methods have further accelerated drug development, transforming modern medicine. However, despite these advances, fewer than 10% of drug candidates succeed in clinical trials (Thomas, D. et al. Clinical Development Success Rates and Contributing Factors 2011–2020 (BIO, QLS & Informa, 2021)). Challenges like pharmacokinetics and the complexity of diseases hamper progress. While we no longer fear smallpox or polio and have effective treatments for bacterial infections and Hepatitis C, today’s most damaging diseases are complex and hard to treat due to our limited understanding of their mechanisms.

Nature 627, S2-S5 (2024) https://doi.org/10.1038/d41586-024-00753-x

Cue paradigm shift. DeepMind’s AlphaFold has revolutionized biology by accurately predicting protein structures, a task crucial for understanding biological functions and disease mechanisms. The economic prowess of Deepmind is also creating some mind-blowing figures. The estimated replacement cost of current Protein Data Bank archival contents (the dataset from which the AlphaFold models were built) exceeds US$20 billion (assuming an average cost of US$100,000 for regenerating each of the >200,000 experimental structures). AlphaFold has subsequently generated a database of more than 200 million structures. Some back of the envelope maths infers that this would have cost us $20,000,000,000,000 using the original methods.

Number of protein structures in Alphafold. Credit: Deepmind

Of course, there are many simultaneous attempts to move the research needle using AI. A team from AI pharma startup Insilico Medicine, working with researchers at the University of Toronto, took 21 days to create 30,000 designs for molecules that target a protein linked with fibrosis (tissue scarring). They synthesized six of these molecules in the lab and then tested two in cells; the most promising one was tested in mice. The researchers concluded it was potent against the protein and showed “drug-like” qualities. All in all, the process took just 46 days. Scottish spinout Exscientia has developed a clinical pipeline for AI-designed drug candidates.

Not only does the platform generate highly optimized molecules that meet the multiple pharmacology criteria required to enter a compound into a clinical trial, it achieves it in revolutionary timescales, cutting the industry average timeline from 4.5 years to just 12 to 15 months. These companies have the technical know-how to build the models, and most likely some internal data with which to train them on. But they need more.

The Power of Existing Data

Platforms like the Dimensions Knowledge Graph, powered by metaphactory, demonstrate the potential of structured data. With over 32 billion statements, it delivers insights derived from global research and public datasets. Connecting internal knowledge with such vast external data provides a trustworthy, explainable layer for AI algorithms, enhancing their application across the pharma value chain.

Knowledge democratization bridges the gaps in the pharma value chain. Credit: metaphacts

AI is not all there is to be excited about in drug discovery. A further technological, serendipitous paradigm shift could amplify the results of AI alone. Once trained, machine-learning models can be updated as and when more data become available. With ‘federated learning’, separate parties update a shared model using data sets without sharing the underlying data. Advances in federated learning allow for collaborating across organizations without sharing sensitive data, maintaining privacy while pooling diverse datasets. Federated learning is a machine learning technique that allows models to be trained across multiple companies holding local data samples. Instead of sending data to a central server, each device sends its model updates (e.g., weight changes) to the central server. This allows further reduction in time and cost in the drug discovery process by improving predictive models, without leaking private company held datasets. Public data can augment local datasets held in corporate R&D departments, enriching the training process. Public data with similar characteristics can help in creating more comprehensive models. This is why we need more, better described open academic research data.

Pharmaceutical companies of the world should be engaging further with both open academic data aggregators in order to assist in the improvement of metadata quality and highly curated linked datasets like the ones supported by the Dimensions Knowledge Graph and metaphactory. The limiting factor is not the AI capabilities, it is the amount of high-quality, well described data that they can incorporate into their models. They need to:

1. Acquire: Gather data from diverse sources, including internal external datasets. Make use of federated learning.
2. Enhance: Enrich data with metadata and standardized formats to improve utility and interoperability.
3. Analyse: Use new models to establish patterns, trends and drug candidates.

You may be thinking that this isn’t a serendipitous leap. This is the fruition of decades of research moving us to a point where these technologies can be applied. You may be right. Either way, the timing of these paradigm changing tools does feel serendipitous. Without AI and federated learning, we could not tackle today’s complex diseases in such an efficient manner. There is a long way to go, but by continuing to curate and build on top of academic data, we can push the boundaries of what’s possible in modern medicine.

This is part of a Digital Science series on research transformation. Learn about how we’re tracking transformation here.

The post The next serendipitous paradigm shift for drug discovery appeared first on Digital Science.