[Bioc-devel] Flexibility for submitting a new ExperimentData-package

[Marcel Ramos]

Hi Teemu Daniel,

Those packages were developed before ExperimentHub.

Please follow the current standard for experiment data packages at

https://bioconductor.org/packages/release/bioc/vignettes/ExperimentHub/inst/doc/ExperimentHub.html

You may want to take a look at curatedTCGAData as an example 
(disclaimer: I developed the package).
It is advisable to split the resources into more modular components, 
e.g., colData, sampleMap, data assays
rather than distribute serialized MultiAssayExperiment objects.

IMO, there should be no compatibility issues as the changes would only 
affect the underlying code and
not the user interface.

Best regards,

Marcel


On 11/24/22 1:57 AM, Daniel Laajala wrote:
> Dear Bioconductor devs,
>
> thanks for all the hard work. Regarding if there's any leeway in how ExperimentData-packages are structured:
>
> We started working a few years back on a "curatedPCaData" R-package for curating prostate cancer multi-'omics datasets. We followed the structure of a family of similar packages, e.g. readily on bioc:
> https://bioconductor.org/packages/release/data/experiment/html/curatedBreastData.html
> https://bioconductor.org/packages/release/data/experiment/html/curatedOvarianData.html
>
> With the data in a satisfactory form, we were now hoping to submit the package to Bioconductor. However, due to the size of MultiAssayExperiment-objects, the package is auto-rejected:
> https://github.com/Bioconductor/Contributions/issues/2804
>
> The data-raw/-files we can discard, as we provided these just for extra user convenience.
> However, the mae_*.rda files are essential MultiAssayExperiment objects (as a standard format for multi-'omics in Bioc), but these are right now too large for the bot.
>
> Reading the package structure instructions, we were under the impression that this old structure for ExperimentData would still be ok for Bioconductor; the package passed checks (R CMD check and BiocCheck) with minor NOTEs. Our current GitHub is at:
> https://github.com/Syksy/curatedPCaData
>
> Our package already has citations (with code provided there-in for extracting and processing relevant data from the package), has been forked, has collaborators using it as-is, etc, so revamping it to e.g. ExperimentHub-structure would most likely cause quite major compatibility issues.
>
>
> Would there be any chance of approaching Bioconductor with the package roughly in its current format inspired by these older curated-packages, even if it is not in the updated ideal ExperimentData-format?
>
> Kind regards,
> Teemu Daniel Laajala
>
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> Bioc-devel using r-project.org  mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel

This email message may contain legally privileged and/or...{{dropped:4}}