You are here : Version Anglaise > Open Science

Open Science

The publication of an article in an international peer-reviewed journal is a condition for the authorization of the thesis defense (see the precise rules on the page dedicated to the preparation of the defense).

More generally, the doctoral school is committed to promoting "Open Science" practices, in accordance with the second national plan for Open Science and the policies of the supervising institutions in terms of Open Science (see here: Lyon 1 and Lyon 2).

The "Open Science" approach of the NSCo doctoral school is based on two axes:
i) mandatory deposit of a version of the publication manuscript on an open archive, in application of the editorial policy of the journal (submitted/accepted/published version depending on the case - to be checked on the Sherpa Romeo website);
ii) Encouragement of the FAIR principles (Findable, Accessible, Interoperable, Reusable) by providing data, analysis codes, protocol records, etc.

This page proposes a short FAQ about the "open science" criteria required to obtain the authorization to defend at the NsCo doctoral school. It answers some general questions that can be asked for each of the 2 options.

How do I put my manuscript in the open access ?

There are two possibilities :

• the journal in which the article is published is an open access journal: in this case there is nothing more to do, the article will be accessible by anyone interested. A directory of open access journals is available here :

• the article is not yet published or the journal is not open access: in this case, it is possible to deposit the manuscript (before publication!) on a public online archive such as arXiv, BiorXiv, PsyArXiv, MedRxiv, or on the French institutional site HAL.

How many does it cost ?

There too it depends on the site :

• on public or institutional archive sites: no cost

• for open access journals: some are free (Diamond model, for example PeerCommunity), but others charge an article processing charge (APC). The DOAJ directory allows you to filter newspapers by this criterion. Some subscription journals (especially among the most reputable ones) operate with a "hybrid" system: they are not "open" but charge an additional fee to authors to put a new article in "open access". Choosing this solution means paying twice (subscription and open access), and it is then preferable to put the "author-accepted manuscript" on an open archive (HAL, BiorXiv, ...) knowing that the journal sometimes imposes an embargo which cannot exceed 6 months (but see here how to avoid this embargo).

• note that some publishers sign "transformative agreements" with institutions to lower the cost of publication per article (which is in fact directly paid by the institution). For example this agreement between Couperin (a consortium of French academic and research institutions) and the publisher Wiley. This type of agreement is not yet very frequent but it is important to watch out for them.

Won't my thesis work be devalued if it is published in open access ?

A common fear of open-access publishing is that these journals are less selective and demanding on the content of articles and that their content can be less trusted. However, while there are indeed so-called "predatory" journals whose sole purpose is to attract people willing to pay to be published, there are also open access journals that do a very good job of peer review. Some of them are even of very high quality like PLOS Biology or eLife. Whatever the journal, it is up to the author to pay attention to the reputation of the journal and to what he/she generally finds there (B!son is an example of a tool that helps to choose a suitable and quality journal). Note that, if one does not wish to restrict oneself to open-access journals, the dissemination of the manuscript before publication on open archives remains a simple and efficient way to disseminate one's work to all (see above).

What a registered report ?

A registered report is an article that is submitted with only the introduction and methods. The peer-reviewing process will therefore only focus on its scientific interest and the expected quality of the experiments and analyses. Beware, this requires a lot of work to describe in great detail the protocol and the analyses and the statistics that will be done. Once the registered report has been accepted, the experiments and analyses must be carried out, and after verification that the experimental plan has been followed, the journal will publish the results whatever they are, without any other request. It is therefore much easier to publish results that turn out to be negative with this type of article. It should be noted that registered reports are nowadays more adapted to experiments that have precise hypotheses to be tested, but that an unexpected observation during the experiment can be published in the registered report in a special section that specifies that the experimental plan is not being followed. Here are some examples of generalist journals that offer this type of publication : Nature Communications, PLOS Biology, eNeuro, PLOS One, Scientific Reports (more complete list here)

What is a data paper ?

A data paper is a peer-reviewed article, published in a traditional journal, that describes a dataset that is generally open access. Formatting data, cleaning them and providing exhaustive metadata requires a substantial effort that is then valued by this type of publication. Anyone who uses the data later can cite the data paper.
Here are some examples of journals offering this type of publication : Scientific Data, GigaScience (more complete list here)
To go further :

How to learn more about open-science ?

• On the government website : the Open science Passeport, a simple and concise guide for PhD students, and more generally the website Ouvrir la science
• Another FAQ on open access publications here
• Another FAQ on preprints here
• Another FAQ on registered reports here
• Another FAQ on data paper here
• An updated list of likely "predatory" journals : Beall’s list