Computational chemists welcome ‘living’ journal

For computational chemists, there is a broad landscape of software and techniques they can use to model the properties and behaviors of molecules and reactions. Picking which software to use and figuring out how to use it can be time-consuming and complex. A new journal wants to help researchers navigate those challenges by providing a space for scientists to publish educational material such as tutorials and reviews.

David L. Mobley of the University of California, Irvine, Michael R. Shirts of the University of Colorado, and Daniel M. Zuckerman of Oregon Health & Science University came up with the open access, online-only Living Journal of Computational Molecular Science. While some computational chemists currently write up tutorials and best practices for use within their groups or between collaborators, this journal hopes to share those insights with a wider audience.

The journal gives each paper a digital object identifier and authors keep a version on a website called GitHub, which also allows readers to comment on the paper. Authors can update and revise a version of the published paper on that repository. And if they feel like they have made significant changes, the authors can resubmit a paper to be published as a new version, with a new DOI. Existing journals, like F1000 Faculty Reviews, encourage authors to publish revised articles but Mobley says they don’t allow for smaller, incremental updates as the new journal does.

Mobley says he and the other editors created the journal to incentivize computational scientists to create educational material for the field. Colleagues, potential employers, and others measure researchers’ success in part by the number of papers they publish, he points out. Through this new journal, computational scientists can add to their publication count while creating educational material.

Justin Lemkul, a computational biochemist at Virginia Tech, is one of many scientists who have written tutorials for molecular simulation programs. He has long hosted those on his own or a colleague’s website, but now he has published a tutorial on the program GROMACS in the journal’s first issue. Lemkul says he gets a citable publication for his work and others know the tutorials have been vetted through a peer-review process. Computational chemist Angela K. Wilson of Michigan State University says it’s good that people can get credit for work they had been doing for free, though she says it remains to be seen how hiring and tenure committees will weigh publications in the journal compared with those in more traditional outlets.

The journal’s first issue is scheduled to appear Jan. 9, but six papers are already online. Two are tutorials about using specific simulation and modeling software and four are best practices articles describing how to simulate things like lipid bilayers. Mobley says other categories will include reviews, comparisons of different software packages, and articles describing lessons learned from unsuccessful experiments. “The field is moving rapidly, and having sources that are always up to date could be very beneficial,” theoretical chemist David Sherrill of the Georgia Institute of Technology says.

The journal has a few other quirks. Authors retain the copyright of their work, only licensing it for the journal to use. Submitted papers go through a standard peer-review process but one reviewer must be a graduate student or post doc to ensure that the articles are understandable to a more junior audience. Also, Mobley says, grad students and post docs are often working more closely with software than principal investigators and might catch things senior researchers wouldn’t. One thing the journal won’t do is publish new results, Mobley says, unless someone wants to compare a new piece of software to existing ones.