Which details of the cellprofiler analysis should be included in a paper?

Usage & Issues
#1

Hi everybody!

I read the instructions on how to cite CellProfiler and I found in some papers that usually the pipeline the authors used is briefly described in the methods.
I wonder which details should be included concerning the pipeline that was used to obtain the data described in the paper.

In particular, in my case, image analysis will be just a really small section of a part of the paper. The pipeline I used identifies nuclei and cells and calculates mean signal intensity from intracellular staining, after adjusting for background intensity (details here).
Should I just briefly describe what the pipeline does? If yes, should I do so in the methods or in supplementary materials? Is it necessary to include details such as smoothing factor or threshold intensity? If anyone is aware of any paper using CellProfiler for similar purposes, would her/him be so kind as to send me the link, so that I might use it as an example?

Thank you so much for your time!
best!
LFS

#2

Hello LFS -

Let me offer my opinion from the perspective of publishing scientific
research. This is not specific to CellProfiler, nor any particular
journal, nor even biology as a field.

What you should do is make your CellProfiler pipeline available
to any readers who want to understand it in detail and/or use it
for their own work.

(I would also include a brief description of the pipeline, of length
and level of detail appropriate to the journal and overall paper, in
the actual paper.)

Does your target journal offer (or require you to use) a
“supplementary materials” facility where you can submit data
and/or code addenda to you paper that the journal doesn’t
print in its dead-trees version, but does archive and make
available to its subscribers (or, better still, the public at large)?

If so, use that, and include your CellProfiler pipeline, together
with any supporting code that is reasonably needed to run the
pipeline.

In addition (or if your journal doesn’t offer this) place your CellProfiler
pipeline is some (hopefully) archival, publicly accessible repository,
and include a link to it in your paper. Does your institution have
offer any de facto archival repositories? Maybe something like
github would work (although who knows where that code will be
twenty years form now).

(If your journal won’t let you include a link in your paper, argue
with them, and consider publishing in a less-obstructionist
journal.)

I understand that what I suggest is largely not the current norm.
Some people might view this as tactically disadvantageous.
(Someone might see my code, warts and all! Someone might find
a mistake! Someone might use the code for their own research
without having to rewrite it from scratch, and get ahead of me!)
Do it anyway.

Thanks, mm

3 Likes