5

A while ago, I spent some time working for a lab doing biology research. While I was there, my main job was to write code to automatically perform specific kinds of analysis on the collected data, because they collected too much data to be analyzed manually. I did that job for a couple months, and then moved on to doing other things.

Now, I'm seeing that it would be quite beneficial to be able to point to a paper that exists with my name on it. Naturally, I'd like it if I could get my name on papers which used that code (because who doesn't want to get more credit for their work). However, I don't know what the standard is for biology papers, and I'd like to know what the common practice is before I have this discussion. Also note that I haven't had any discussions about this topic with the professor who heads that lab or anyone else in the lab, so I don't know what their thoughts are.

Is it standard practice to include on the list of authors of a paper, someone whose only contribution to the lab was the creation of a computer program to automatically analyze the data?

8

The Council of Science Editors has published a White Paper on Promoting Integrity in Scientific Journal Publications that gives a summary of research on authorship and attribution in scientific journals (see section 2.2 of the report). I recommend you begin by reading this material, to get an idea of the general principles for authorship and the required contributions. One of the things that the white paper is designed to address is the practice of "ghost authorship" where a contributor fails to receive authorship credit despite making a substantial contribution to a paper. It also recommends that if someone contributes to a paper, but not enough to receive authorship, then they should still be listed by name in the "Acknowledgements" section.

There is no universally agreed set of rules for the requirements for authorship, but the general view is that a person should be list as an author on a paper if they have made a "substantial contribution" to the paper. You can also find a set of recommended principles for authorship (in the context of medical journals) published by the ICMJE:

The ICMJE recommends that authorship be based on the following 4 criteria:

  • Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND

  • Drafting the work or revising it critically for important intellectual content; AND

  • Final approval of the version to be published; AND

  • Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

The above principles are written by a well-reputed body of medical journal editors. I do not have any source material from editors of biology journals, but hopefully there is a reasonable correspondence in approaches. In any case, whether or not your coding work constitutes a "significant contribution" to the work depends largely on the magnitude of the contribution relative to the overall paper, whether any customisation or innovation was required in the coding, and whether you made a contribution to the actual design of the study you were coding, or just coded a set of methods/procedures that were already determined by others. On its own, coding the analysis for a study would not usually constitute a "significant contribution", but it would usually constitute enough of a contribution to warrant acknowledgment. (Although it is for a different kind of contribution, you might find it helpful to read Parker and Berman (1998), which sets out recommendations for authorship credit for statisticians working on applied research. This is a different situation, but it may have parallels to someone doing coding work on a research project.)

The main problem you will encounter in this case is that you have not negotiated any credit on the paper prior to doing the work. In future, when you are asked to contribute in some way to an academic paper, it is a good idea to have an initial discussion with the principal author, to make agreement on what your contribution will be, and what acknowledgement (if any) you will receive. You can certainly try to negotiate this post hoc but you might find this more difficult, depending on the attitude of the main authors of the studies in question. In any case, if you feel that your contribution warrants a formal acknowledgement in a paper (or even co-authorship), I recommend you contact the principal author of the paper to start a discussion about that. The best way to do this would be to frame your request in the context of the authorship principles set out in the above documents (or other similar studies), so that you can point to an external objective source to back up your request. In the absence of some more direct document, the ICMJE principles should be applicable to the world of biology papers.

3

Others already told you: when a paper is written by that lab that uses your program, it will depend on how important an intellectual contribution you added to the subject of the paper. This comes in two "sizes": authorship for significant intellectual contribution or acknowledgement if your contribution was "merely" technical, i.e. you implemented the analysis someone else developed.

(And it's notoriously difficultimpossible for us to judge whether your contribution would merit authorship or only an acknowledgement without far more details about the project.)


However, there's a third option that has not been mentioned so far: citing your software:

it would be quite beneficial to be able to point to a paper that exists with my name on it.

Possibly too late for the project in question now, but something to keep in mind for future projects: You can have software publications.

  • for software that in itself adds/implements new scientific knowledge that can be scientific paper which is published together with the software.
  • but it may also be the manual of that software (which in science terms would count as a non-peer-reviewed publication similar to e.g. a technical report)
  • you can even get a DOI for the software itself.

The first and third options of course requires the software to be public to some extent - so this would be something to discuss with the PI of the project.

Naturally, I'd like it if I could get my name on papers which used that code (because who doesn't want to get more credit for their work).

You can then ask that this publication be cited by papers that used it.

This is quite common e.g. in the R world. Actually so common that there is even a function citation() that tells you how each package would like to be cited. (see also e.g.here for reasons to cite scientific software).

Here are some examples of packages I am/was involved in that illustrate the range of possibilities:

  • package softclassval is an implementation that belongs to a full scientific paper:

    > citation ("softclassval")

    To cite package 'softclassval' please use:

    softclassval-ChemomIntellLabSyst

    Claudia Beleites, Reiner Salzer and Valter Sergo: 'Validation of Soft Classification Models using Partial Class Memberships: An Extended Concept of Sensitivity & Co. applied to grading of astrocytoma tissues', Chemometrics and Intelligent Laboratory Systems, 122 (2013), 12 - 22, DOI: 10.1016/j.chemolab.2012.12.003, arXiv: 1301.0264, R package version 1.0-20160527, http://softclassval.r-forge.r-project.org.

  • package hyperSpec is publicly available on CRAN and github, the citation refers to the package/its manual:

    > citation ("hyperSpec")

    To cite package 'hyperSpec' please use:

    Claudia Beleites and Valter Sergo: `hyperSpec: a package to handle hyperspectral data sets in R', R package version 0.99-20200114.
    https://github.com/cbeleites/hyperSpec

  • package cbmodels is a collection of functions I put together when I was working at IPHT. It is not under an open source license (it's under the "ask-the-institute-director license"). I don't have it any more, so I cannot call citation() on it, but we used e.g.

     C. Beleites, cbmodels, R package version 0.5, 2015.

    in a publication.

Note that version numbers also serve scientific reproducibility.

0

@reinstate-monica has already pointed out, these discussion are best happening before the work is done. However, for what its worth if you were in my lab, you would absolutely be an author on the paper. BUT .... this does mean would also expect you to be intellectually engaged with the project and prepare to offer feedback on the whole paper (including parts you weren't involved in). But in any case you should definitely be acknowledged. I think the previously mentioned comparison to applied statisticians if perfectly apt.

I think two apt questions are 1) Could the paper have gone ahead without your contribution, 2) If someone were to challenage the correctness of the part of the paper you were involved with, would any of the authors be in a position to know precisely what happened, understand the process and comment on any possible mistakes, omissions and design decisions. Somebody on the the authors list should understand every part of the process sufficiently to take responsibility for it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.