This article describes how LiveCoMS article preparation and submission work, as well as giving details of the review process, revisions, authorship, and other aspects, as follows:
- Submission Process
- Before Submission
- Submission and the Review Process
- Updating LiveCoMS Articles
- Authorship and Changes to Authorship
- Other Policies for Submitted Articles
Submitting an article to LiveCoMS is substantially different from submitting an article to most journals. In this article, we lay out the process by which authors create an article, submit it to LiveCoMS, and update it over time to create a living document.
There are no explicit length limits on any manuscript. Articles do not need to contain original research, but may contain it.
The goal of best practices manuscripts are to improve the quality of simulation research by providing advice that can eliminate the most common errors. These documents consist of:
A. A checklist: A succinct list of steps that people should follow when carrying out the task in question. This is effectively a checklist, provided to ensure certain basic standards are followed and common but critical major errors are avoided.
B. The rationale: A much more detailed section with the necessary rationale for the checklist, which would act as more complete best practices description, with significant detail as to the possible alternative ways to accomplish a given task, when one would be better than another, and significant literature documentation about reasons for choices.
The inspiration for Best Practices documents can be found in the Atul Gawande article The Checklist, later expanded in his bestseller The Checklist Manifesto. The main goal of these documents is to reduce common errors by ensuring appropriate choices are made throughout a difficult task.
For our purposes, simulation checklists should help users avoid the most common reasons for failure or incorrect results. Checklists will typically also be accompanied by a detailed explanation with sources and references (the rationale mentioned above), and this rationale can go into more detail on best practices and cover more possible failure modes and how to avoid them.
One can divide the types of errors that are made performing molecular simulations into (a) mistakes experienced researchers often make, and (b) mistakes new users often make, even after having received solid training in fundamentals of molecular simulation. Most checklists will be focused on these types of failures. Potentially, authors might also be interested in helping to prevent the mistakes of naive users who do not understand the basic principles of molecular simulation. However, normal best practices documents can assume a basic familiarity with the fundamentals of molecular simulations; a separate class of documents can be targeted at introducing novice users to the field, and these will serve as prerequisites for more detailed documents.
Perpetual reviews, like standard reviews, seek to provide an overview of an area of the field; however, they can be updated as needed, thus serving as living documents that remain current. Perpetual reviews change both in response to community input, and in response to new developments in the field. An example of this process can be found in the Mobley et al. review, which was originally an Annual Reviews in Biophysics article, but is now being updated regularly on GitHub.
Simulation comparison papers describe attempts to perform the same calculation with a range of different simulation programs. Such comparison should be updated periodically with different versions of the same programs (or potentially additional programs).
Tutorials are articles that come accompanied by web pages with walkthroughs or instructional materials and downloadable files to work with; e.g. an explanatory article accompanied by a website with training information and downloadable Jupyter notebooks teaching users how to perform a challenging modeling task may be an ideal tutorial.
Tutorials take users through specific tasks or sequences of tasks to help them learn how to do these tasks on their own for other applications or purposes.
Scope of tutorials: Tutorials should endeavor to not just cover the specific task at hand, but also highlight how the steps might need to be modified (or additional care might need to be taken at particular points) to handle more general cases. Tutorials for specific software packages must provide instructions and files for the current version of the software.
Pre-existing tutorials: The submission of existing tutorials, so long as they meet journal standards, is explicitly welcomed. Our goal is to encourage the highest quality tutorials by providing some degree of academic credit for these important and time-consuming efforts.
Authors should first send a presubmission letter to the lead editor in the relevant area. The letter should be no more than one page, and should:
- Outline the scope of their best practice document or review.
- Explain how the proposed manuscript is different from existing best practices articles or reviews.
- Note whether it is adapted from a previous article (and identify the article if so).
- Explain the expertise that the proposed authors have on the subject.
- Describe the license that the authors will use that will enable the article to be released freely to the public.
Potential authors submitting a presubmission letter will typically receive an answer within two weeks with either encouragement for a full submission, suggestion to work with previous authors on existing articles, or discouragement. Articles should be received within six months of a notification of encouragement.
The authors should prepare the document using LaTeX in a public repository owned by one of the authors (or their organization/group) on GitHub. LiveCoMS provides template LaTeX files to start from, and instructions for exactly how to structure the documents. We require that articles that are submitted to LiveCoMS use the provided templates for the articles so that the journal has a consistent visual presentation. Additionally, articles should include clear links to and mention of the relevant GitHub repository and encourage community participation/feedback via GitHub.
For an example of an article hosted on GitHub in this style (though not a LiveCoMS article), see the following perpetual review.
One written, the authors post the resulting PDF document to a preprint server of their choice, which can be any one of:
- ArXiv
- BioRxiv
- ChemRxiv
- Faculty of 1000 Research
- Any of the Open Science Framework preprint servers (engrXiv, etc.)
Posting to the preprint server can be done at any time prior to submission.
After posting to the preprint server, the author uploads the final article PDF (created using the LiveCoMS templates, discussed above) and a link to the GitHub site for the article.
The authors must also submit a cover letter with suggestions for 4--5 reviewers, and note any deviations from the presubmission letter.
The editors will choose reviewers, which may include the editors themselves.
Reviews will generally be anonymous, though reviewers will be allowed to become known if they desire, and can participate directly in revisions through the GitHub website, whether or not they remain anonymous.
For example, a reviewer could choose to submit a very brief review addressing only suitability, but provide extensive feedback to the authors on the GitHub issue tracker, allowing discussion of how the article should be revised to be done openly.
This open revision approach may be particularly suitable for articles which will become community resources.
A key purpose of the articles is that they should be useful to a range of researchers, but especially beginning researchers. Thus, all best practices and review articles will be reviewed by a member of the student review board, which consists of graduate students and postdocs invited by the editorial board.
Authors are also encouraged to have other researchers review their content, with comments and responses handled via the article's GitHub issue tracker. A history of revisions in response to community concerns will impact the review process favorably.
Reviews will also consider the following factors according to manuscript category:
-
Best practices guides
- Would following the provided checklist help users avoid significant potential errors in simulation? Will the errors be profound and/or frequent?
- Are all assertions well-sourced in published data (which may, in some cases, include data created for the paper)?
- Are the explanations clear enough for researchers with only moderate training in simulation?
-
Perpetual reviews
- Do the authors properly include information from the recent literature, including last 6 to 12 months?
- Is this a good summary of its area which will be valuable to the field?
- Do the authors include data and viewpoints that are contradictory to their own, if these exist?
- (If a revision) Have the authors removed data that is out of date, and qualified information that has been disproven or contradicted?
-
Comparisons of molecular simulation programs
- Does the paper include developers or advanced users of all programs being compared to ensure that comparisons are fair?
- Are best practices being used in the simulation comparisons?
- Have the authors addressed whether a naive user could expect similar performance and why or why not?
-
Tutorials:
- Are files to run the simulations posted online in a permanent (or nearly so) way?
- Are the tutorials helpful and educational?
- Do they include enough information on how to generalize the approach to handle other cases, and highlight major issues to consider?
- Do the tutorials use commonly agreed on best practices?
If manuscript revision is requested, authors will be asked to re-submit within 14 days for minor revisions and 30 days for major revisions. After this time, a revised manuscript may be handled as a new submission.
A unique aspect of LiveCoMS is article versions, where new, updated versions of articles can be re-reviewed and treated as new publications. By using GitHub to write the paper, questions, comments, or additions from the community can be filed as issues and then can be easily incorporated into new versions of the article. In this way the document becomes truly a living document.
Once peer reviewed, articles receive new DOI’s and are published on the LiveCoMS site as new versions. This allows authors to receive credit for ongoing work they do on their articles.
Authors are encouraged to make updates to articles in their GitHub repositories as frequently as warranted. However, release of new peer-reviewed versions via LiveCoMS is warranted only when changes become particularly extensive or important. Thus, versioning should typically be done no more frequently than every 12 months.
The review process for an update to a LiveCoMS article is similar to the review process for the initial version. Additional review criteria will include whether or not issues the community raised on the article's issue tracker were responded to, and whether the revision includes sufficient new material. Certain categories of article may need revision at different frequencies:
-
Best practices guides: These articles may start with more frequent versions, as initial rounds of community input are solicited. As always, authors are encouraged to update their private GitHub versions as frequently as desired, and if the document changes rapidly, more frequent new versions on LiveCoMS may be needed; consultation with your editor is encouraged.
-
Perpetual Reviews: Authors will be asked to update reviews at least once every 36 months, and will be encouraged to ensure continuity of authorship (e.g. if one author retires or loses interest) so the review can be maintained, or else the review will be “made emeritus” and indicated as out-of-date on the site. We recommend updating at least every 24 months. However, authors decide when it is time to re-version based on feedback they receive.
-
Tutorials: Authors should generally update tutorials as the software involved changes; these tutorials are expected to be kept functioning and if they do not, they may be indicated as out-of-date on LiveCoMS' site.
Authorship for "living" documents, LiveCoMS' focus, can become complicated as the document involves and more people contribute. Our key principle is that participants should get credit for their contributions, which can come in a variety of forms such as actually writing the work, providing feedback and filing issues, etc. However, different types of credit may be warranted. In general, changes which constitute writing a significant part of the article merit authorship, but not those which only modify small portions. These smaller changes, however, should still be acknowledged.
In order to acknowledge these contributors, people who offer comments/citations that are used in the paper should be listed listed on the relevant GitHub repository README Markdown file, and should be listed in the acknowledgment section of the paper. However, if the current authors feel that the contributions rise to the level of authorship, they can add new authors when the next major version is submitted.
Exactly what constitutes a "significant" contribution is by necessity subjective, and authors should endeavor to be generous. In the process of reviewing revisions, the editors may examine the history of the GitHub repository to ensure appropriate credit is being given. If contributors feel that they should be given authorship based on the number of accepted commits, the editors will help resolve any disputes. If contributors feel that their commits are not being accepted by the authors they should first strive to convince the authors and the community through well documented, convincing arguments. If authors do not engage with substantial issues that have been raised, they may be asked to address these issues in the review process. If authors do not substantively engage in the discussions on issues that are raised, this may prevent new updates of the review from being accepted.
Authors are required to report funding sources and grant/award numbers relevant to the manuscript for ALL authors.
Authors should note whether any funding could be perceived as a conflict of interest, e.g., an article describing software in which an author has a financial interest.
LiveCoMS does not require any copyright transfer to the journal.
Instead, LiveCoMS requires articles use a license which allows LiveCoMS to distribute the articles freely consistent with Open Access requirements.
We recommend that all submissions to LiveCoMS be released by the author under a Creative Commons By Attribution license version 4.0; however, other licenses may be acceptable with justification.
Documents should not have been submitted in the current form to another journal, or be simultaneously under consideration for publication another journal. Preprints do not count as prior publication. Documents that are major revisions of previously published articles are welcomed. However, authors should ensure that material they develop for the journal does not have rights reserved by other journals from such previous publication that impede release under open licenses. Some journals, for example, Annual Reviews lets the authors retain the right to create derivative works, which could perhaps be exercised in preparing a review to be published in LiveCoMS.
If an article is an adaptation of a previously published article, it must be noted in the cover letter, and major changes noted. Evaluating whether such changes constitute a significant revision will be part of the review process.