3

I am a PhD student in computer science. Together with my advisor we are collaborating with some external centers, specialized in biology research;

the idea is that I build an algorithm, we use this algorithm on their data and then we publish the results.

Now, we have found some pretty interesting results on a set of their data and we are planning to write a paper; however the target of the paper will be mainly a biology review, not a cs conference.

Even if we are going to publish the results on a prestigious review, I would like also to write a paper more focused on the algorithm, maybe for a conference of applied computer science, using the same data; consider that a paper with only the algorithm and the relative details has already be submitted.

Does it make sense? Will a publication on a top review for a different field be considered in any way when I will apply for future post doc positions, or is it better to write a second paper, more focused on my field?

  • 2
    From a technical perspective, it makes sense that you can present a paper about the actual data, and one about the implementation of the algorithm. You will need to cite the original paper in your CS paper to point to the practical analysis of the data from the algorithm, and make sure to talk to your advisor about it, but I don't see why a technical paper is out of the question. – Compass Dec 15 '16 at 19:18
6

It is entirely reasonable to produce more than one paper from the same project, focusing on different aspects of the the project. Legitimately complicated projects often produce many papers worth of results: it only becomes unethical salami slicing if you're trying to "stretch" one paper into many.

The type of situation that you describe, however, is an extremely common circumstance for ending up with two papers, one focused on method (in this case, your algorithm) and the other focused on results (in this case, what has been learned from the biological data). The two papers will necessarily be quite different: the methods paper must investigate questions of scaling, scope of applicability, etc., far beyond this particular dataset. Complementarily, the applications paper needs to consider underlying theories, alternative explanations, hypotheses, etc. with little bearing on how the data happens to have been analyzed.

From my own personal experience as both an author and a reviewer, I would recommend doing this in one of two ways:

  • If the two have similar audiences, you can co-submit two papers that reference one another to the same journal. Explain this in the letters to the editor, and they will likely receive special handling that considers them as a linked unit, e.g., the same handling editor, possibly shared reviewers.

  • If the two have very different audiences, publish whichever is ready to go first, mentioning the issues of the other only briefly, then follow up with the other citing it. This can be in either order, though if you're dealing with a high-impact biology journal, it may be wise to go with the biology paper first, since such publishers are sometimes touchy about "novelty" concerns.

| improve this answer | |

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.