Tag Archives: Wellcome Trust

Open Research Project, first thoughts

Dr Laurent Gatto is one of the participants in the Office of Scholarly Communication’s Open Research Pilot. He has recently blogged about his first impressions of the pilot. With his permission we have re-blogged it here.

I am proud to be one of the participants in the Wellcome Trust Open Research Project (and here). The call was initially opened in December 2016 and was pitched like this:

Are you in favour of more transparency in research? Are you concerned about research reproducibility? Would you like to get better recognition and credit for all outputs of your research process? Would you like to open up your research and make it more available to others?

If you responded ‘yes’ to any of these questions, we would like to invite you to participate in the Open Research Pilot Project, organised jointly by the Open Research team at the Wellcome Trust and theOffice of Scholarly Communication at the University of Cambridge.

This of course sounded like a great initiative for me and I promptly filed an application.

We had our kick-off meeting on the 27th January, with the aim of getting to know each other and somehow define/clarify some of the objectives of the project. This post summarises my take on it.

Here’s how I introduced myself.

Who are you?

Laurent Gatto, Senior Research Associate in the Department of Biochemistry, physically located in Systems Biology and the Maths Department. SSI fellow and Software/Data Carpentry instructor and generally involved in the Open community in Cambridge, such as OpenConCam and Data Champions initiative.

What is your research about and what kind of data does your research generate?

My area of research is computational biology, with special focus on high-throughput proteomics and integration of different data and annotations. I use raw data produced by third parties, in particular the Cambridge Centre for Proteomics (mass spectrometry data), and produce processed/annotated/interactive data and a lot of software (and also here).

What motivated you to participate in the Pilot?

Improve openness/transparency (and hence reproducibility/rigour) in my research and communication, and participate in improving openness (and hence reproducibility/rigour) more widely.

What kind of outputs are you planning to share? Do you foresee any difficulties in sharing?

My direct outputs are systematically shared openly early on: open source software (before publication), pre-prints, improved data (as data packages). Difficulties, if any, generally stem from collaborators less willing to share early and openly.

A personal take on the project

It is a long project, 2 years, and hence a rather ambitious one, of a unique kind. Hence, we will have to define its overall goals as we go. The continued involvement of the participants over time will play a major role in the project’s success.

What are attainable goals?

It is important to note that there is no funding for the participants. We are driven by a desire to be open, benefit from being open and the visibility that we can gain through the project, and the prospect that the Wellcome Trust will learn from our experience and, implement any lessons learnt. We get to interact with each other and with research support librarians, who will help us throughout the duration of the project. We also commit to sharing of research outputs beyond traditional publications and to engage with the Project, by participating in Project meetings and contributing to Project publications.

A lot of our initial discussions centred around rewards for open research or, actually, lack thereof and perceived associated risks. Indeed, the traditional academic rewarding system and the competitiveness in research leaves little room for reproducibility and openness. It is, I believe, all participants hope that this project will benefit us, in some form or another.

A critical point that is missing is the academic promotion of open research and open researcher, as a way to promote a more rigorous and sound research process and tackle the reproducibility crisis. What should the incentives be? How to make sure that the next generation of academics genuinely value openness and transparency as a foundation of rigorous research?

Some desired outputs

Ideally, I would like that the Wellcome Trust’s famous Research investigator awards to be de facto Open research investigator awards. There’s currently a split (opposition?) between doing research and supporting open science when doing research. In every grant I have written, I had to demonstrate that the team had a track record, or was in a good position to successfully pursue to proposed project. Well, how about demonstrating a track record in being good in opening and sharing science outputs? Every researcher submitting a grant should convincingly demonstrate that they are, have been and/or will be proactive open researcher and openly disseminate all the outputs. By leading by example in the frame of this Open Research Project, this is something that the Wellcome Trust could take away from.

Unfortunately, it is a fact that open science is not on the agenda of many (most?) more senior researchers and that they are neither in a position to be open nor that open science is a priority at all. I find it particularly disheartening that many senior academics (i.e. those that will sit on the panel deciding if I’m worth my next job) consider investing time in open science and the promotion of open science as time wasted of actually doing research. A bit like time for outreach and promotion of science to the wider public is sometimes looked down at, as not being the real stuff.

Another desire is that this project will enable us to influence funders, such as the Wellcome Trust, of course, but also more widely the research councils.

As a concrete example, I would like all grants that are accepted to be openly published beyond the daft layman summary. Published grants after acceptance should include data management plan, the pathway to impact, possibly more, and these could then be used to assess to what extend the project delivered as promised.

This serves at least two purposes. First, it is a way to promote transparency and accountability towards the funder, scientific community and public. Also, it is a great resource for early career researchers. Unless there is specific support in place, writing a first grant is not an easy job, especially given the multitude documents to prepare in addition to the scientific case for support. And even for more experienced researchers, it can’t harm to explore different approaches to grant writing.

Another concrete output is the requirement for a dedicated software management plan for each grant that involves any software development. I certainly consider my software to be equivalent to data and document it as such in my DMPs, but there seems to be a need for clarification.

I believe that I do a pretty decent job in conducting open science: pre-prints, open access, release data, … In the frame of this project, I shall do a better job at promoting open science for its own sake.

I also hope that by bringing some of my projects under the umbrella of the the Open Research Project, I will benefit from a broader dissemination that will, directly or indirectly, be beneficial for my career (see the importance of benefits and rewards above).

Next steps

It is important to make the most out of this unique opportunity. We need to create a momentum, define ambitious goals, and work hard to reach them. But I also think that it is important to get as much input as possible from the community. Nothing beats collective intelligence for such open-ended projects, in particular for open projects.

So please, do not hesitate to comment, discuss on twitter or elsewhere, or email me directly if you have ideas you would like to promote and or discuss.

Published 08 March 2017
Written by Dr Laurent Gatto
Creative Commons License

We are going OPEN – the Open Research experiment has begun!

There has been much discussion recently about the reproducibility crisis and about the growing distrust among the public in the quality of research. As illustrated in our ‘Case for Open Research’ series of blog posts, one of the main reasons for this is that researchers are currently rewarded for the number of papers they publish in high impact factor journals, and not necessarily for the quality of work that they are doing.

Indeed, Cambridge researchers clearly indicated that the lack of incentives to do anything other than publishing in these types of journals is one of the main blockers discouraging them from adopting a more open research practice.

Joining forces with the Wellcome Trust

The Office of Scholarly Communication started talking about these problem with the Open Research team at the Wellcome Trust. The Wellcome Trust are natural allies, as they have consistently led their researchers towards greater openness. They were one of the first funding bodies to introduce policies on Open Access and on data management and sharing. Now the Wellcome Trust is moving towards proactively supporting Open Research beyond enforcing their compliance requirements.

To promote immediate and transparent research sharing, they have recently launched the Wellcome Open Research platform which allows researchers to submit articles about virtually any research output and get published within a couple of days. The Wellcome Trust is now considering making Open Research one of their strategic priorities.

We quickly realised that we have a lot of shared interests, and joining forces to tackle the problem together made a lot of sense. We came up with the idea to launch the Open Research Pilot Project.

The Open Research Pilot – understanding the barriers to “openness”

We conceived the project as a two year experiment, which would allow us to gain an understanding of what is needed for researchers to share and get credit for all outputs of the research process. These include non-positive results, protocols, source code, presentations and other research outputs beyond the remit of traditional publications.

The Project aims to understand the barriers preventing researchers from sharing (including resource and time implications), as well as what the incentives are. The Project aims to utilise the new Wellcome Open Research publishing platform, together with other channels, to share these outputs.

The invitation to take part in the Pilot was sent to all researchers at Cambridge funded by the Wellcome Trust. Participating researchers had to commit to sharing of research outputs beyond traditional publications and to engage with the Project, by participating in Project meetings and contributing to Project publications.

Is ‘doing the right thing’ enough incentive?

Our biggest question was whether anyone would be willing to participate in the Pilot. We did not offer any incentive other than encouraging researchers to contribute to the greater good. The only support available to those who wanted to take part in the project was that offered by the Wellcome Trust and Cambridge Open Research team members, but there was no financial aid available to prospective participants. We thought that regardless of the outcome, that inviting researchers would be a good exercise to go through – we thought that if no one applied, we would have learnt that doing ‘the right thing’ was not a good enough motivator.

Thankfully, we received several fantastic applications from individual researchers and research groups who demonstrated great interest in and motivation for Open Research. We initially planned to work with two research groups, but given the quality of applications received and passion for Open Research expressed by the applicants, we decided to extend the scope of the project to four research groups. We have selected researchers doing different types of research, with the aim of learning about distinct problems in sharing that are experienced in diverse research disciplines:

  •       Dr Laurent Gatto –is  doing computational biology research, with a special focus on proteomics data. His interest is: How to effectively share research data and the code needed to reproduce them?
  •       Dr David Savage – is researching molecular pathogenesis of the consequences of obesity. His question is: What are the problems with sharing data coming from human participants?
  •       Dr Benjamin Steventon – is a developmental biologist generating and analysing large-scale imaging datasets. He would like to know: Are there image repositories allowing one to share large image datasets in a re-usable way?
  •       Dr Marta Costa and Dr Greg Jefferis (and others) – researchers leading the work on two collaborative projects: Connectomics and Virtual Fly Brain, which will create interactive tools to interrogate Drosophila neural network connections. They would like to understand: What are the issues with sharing complex interactive datasets? How to ensure long-term preservation of complex digital objects?

Motivations

So what motivated these researchers to apply for the project? We asked this question at the application stage and were positively surprised by the altruistic answers that we received. Our researchers were largely driven by a desire to improve the research process. We have seen responses like:

  • “Openness in research, from data and software to publication, is a central pillar of good research.”
  • “I am very concerned (disappointed as a scientist) by the current wave of ‘unreproducible’ and/or ‘irrelevant’ research, and am very passionate about contributing to improving scientific endeavour in this regard.”
  • I am very enthusiastic about exploiting new ways of sharing my research output beyond the established peer-review journal system.”
  • “I believe that sharing research outputs fully, including data and code are essential to accelerate research, and I have benefitted from it in my own research.”

Summarising, researchers expressed a great desire for contributing to a cultural change. Researchers wanted to change the way in which research was disseminated and to increase research transparency and reproducibility.

Let’s get to work

We all met (the researchers, Wellcome Trust and Cambridge Open Research teams) on Friday 27 January to officially start the two year project. Each research group was appointed a facilitator – a dedicated member of the Cambridge Open Research team to support researchers during the Project. Research groups will meet with their facilitators on a monthly basis in order to discuss shareable research outputs and to decide on best ways to disseminate these outputs. Every six months all project members will meet together to discuss the barriers to sharing discovered and to assess the progress of the Project.

One of the main goals of the Project is to learn what the barriers and incentives are for Open Research and to share these findings with others interested in the subject to inform policy development. Therefore, we will be regularly publishing blog posts on the Unlocking Research blog and on the Wellcome Open Research blog with case studies describing what we have discovered while working together. There will be an update from each research group every six months. We will also be publicly sharing all main outputs of the Project.

We are all extremely excited about going “Open” and we suggest that anyone interested in the Open Research practice watches this space.

Published 08 February 2017
Written by Dr Marta Teperek
Creative Commons License

Are we achieving our OA goals?

This post was written for Hindawi for Open Access Week and published by them on 28 October. It is reposted here.

Recently I spent a day in two consecutive weeks travelling to London to meet with colleagues to discuss the implementation of the Wellcome Trust (COAF) and RCUKOpen Access policies. In both cases the discussions were centred on compliance with their policies.

Certainly it makes sense that a funder should ensure that its policies are being implemented properly. But this focus on compliance raises the more fundamental question about whether we are actually achieving the underlying goal of these policies – which is to open up access to UK research so more people can access, read and use this work.

After all, having huge swathes of research in repositories under embargoes* or spending literally millions of pounds annually to make particular articles in subscription journals available open access is not in itself the end goal.

We should be taking stock. Have the past three and a half years of the RCUK and over a decade of the Wellcome Trust policies meant our researchers are more engaged in open access? Has there been a movement by publishers towards flipping their journals? Indeed, is UK research being read and used more now? These are very pertinent questions that simply do not appear to be discussed at the moment.

*Cambridge has managed to address this issue by providing a Request a Copy button – see here.

Big bucks

There is a lot of money in this ecosystem. Cambridge University has been allocated £1,269,318.59 by RCUK in the 2016-2017 year, and have a £403,138 underspend which will be directed to this year’s Open Access activities. In addition Wellcome Trust have allocated us £902,243.

So Cambridge University has £2,574,699.59 allocated by funders to pay for Open Access APCs and related staff and systems costs (we recently made all of our expenditure available). Cambridge University spends about £4.8 million annually on subscriptions, so the cost for Open Access at our institution is over half of our subscription cost.

These are serious amounts of money. Surely it is a good idea to ask whether this process is actually achieving what it set out to do.

So what has actually happened?

Embargo changes

The RCUK Open Access policy has allowances for green Open Access with a sufficient embargo period and the decision tree at the Office of Scholarly Communication reflects the actual wording and rules of the policy – that is choose green options if you can. However the emphasis of RCUK is decidedly towards gold Open Access – see their decision tree which is actually slightly misleading.

So when the RCUK announces a policy where cash for article processing charges will flow to publishers dependent on embargo periods, what happens? The embargo periods lengthen.

According to a study published this year “What does ‘green’ open access mean? Tracking twelve years of changes to journal publisher self-archiving policies” (Open Access version here) there is “a clear link between the introduction of Gold open access and the increasing restrictions around Green open access”. The study also includes a graph mapping embargo periods over time which shows a very clear and defined ‘Finch effect’.

This was entirely predictable. When the RCUK Open Access policy was announced in response to the Finch Report I wrote (in my previous role) “Clearly it is advantageous for journals to offer a hybrid option and to extend their embargo periods in response to this policy.” And they did.

Springer and Emerald both extended their embargoes beyond the RCUK limits. (Of course Springer has since redeemed itself by experimenting with new business models).

Those embargo extensions were particularly galling at the time for me because they were worldwide and affected everyone – including in Australia, where I was based. Other publishers have responded to the RCUK rules by creating particular embargoes for UK authors. Elsevier is a clear example.

Institutional pressure

About the time the RCUK policy came into force I wrote about the difficulty of anyone staying up to speed on copyright agreements. Since then it has got worse. At Cambridge we do not expect our researchers to try and wade through this – we provide a service to help them. But this means staff and that costs money.

The pressure on research institutions to manage the UK Open Access policies is significant. Analyses of the total cost of publication (Open Access version here) associated with the administration of making research open access show a huge staffing load. The cost of processing a gold Open Access article was shown to be 2.5 times that for the processes of making an article available in a repository.

The RCUK do allow some of their block grant to be spent on staffing and infrastructure. At Cambridge we have reported that we spent 4.6% of the year’s allocation on staff costs and 5.1% on systems support. The general understanding is that RCUK don’t want the total spend on these costs to be more than 10% of the grant and it appears some institutions have spent more than this in previous years.

This highlights the overall lack of funding for support costs for managing Open Access. There are no specific funds for managing the HEFCE Open Access policy, or the COAF policies. While both the Wellcome Trust and HEFCE provide considerable funds to UK institutions for research, this is not directed to the Libraries. Certainly at Cambridge there is a robust process required to argue for funds to support these types of activities.

The 2012 Finch Report talked about a “transition to open access” and acknowledged that this will mean additional costs. Certainly the funders have channelled significantly more funds to publishers through the institutional block grants, and those institutions are having to channel internal resources to support the staff supporting the policies.

But the Finch Report also mentioned “seeking efficiency savings and other reductions in costs from publishers and other intermediaries”. It is safe to say that this has yet to actually occur.

Taking stock

So, more than four years on from the Finch Report, are we any closer to full Open Access? The answer is yes in the UK – because we have poured millions of extra (taxpayers’) pounds into the system. But if the RCUK policy were to end tomorrow, would the publishing landscape be any different? Has any other country in the world followed this model?

And are the Open Access policies achieving their end goal? Is UK research more visible in the world now? Are people actually finding these articles? Is it being read more?

Is anyone even asking these questions? Who is monitoring this? If we don’t ask and measure these parameters we will never know.

What we do know is we have extended embargo periods, forcing funded researchers down the gold Open Access path, which is more expensive to process in terms of staff time. We have spent millions, the majority of which is spent in hybrid journals – which is itself another issue. And there is little if any evidence that publishers are moving towards fully Open Access models.

A glimmer

Unfortunately the discussions held recently about the Wellcome Trust and RCUK policies were solely focused on compliance. This has become the narrative in the Open Access space in the UK and does nothing to help ‘sell’ the idea of Open Access.

Indeed it would be hugely helpful if there were communication about the underlying goals of these policies and whether they are being met. But the lack of monitoring of these goals means we have nothing to say. We can’t communicate what we don’t know about.

There is some hope. At least one publisher is interested in whether this is making a difference. At the Frankfurt Book Fair last week I attended a discussion of the German Serials Interest Group where a colleague from Springer said that Springer is assessing the success or otherwise of the Springer Compact. They had specifically compared the readership of Open Access articles against subscription only articles. According to this work, the percentage of non-institutional affiliated people reading the Open Access articles was dramatically higher than the subscription-only.

This type of information is hugely valuable to Open Access advocates, and I am hoping that Springer will release these findings publicly.

The team at the Office of Scholarly Communication strongly believe that all Cambridge research should be available, and we are working hard towards that goal (recently celebrating 10,000 submissions to the repository). It would help us enormously if we could offer evidence to our community of the value and benefits of this effort.

Published 3 November 2016
Written by Dr Danny Kingsley

Creative Commons License