Is it going to be available on the Bioconductor released version?

[tghosh30]

Hi,

Congratulations for writing such a wonderful compact package. I was wondering if this package will be available on Bioconductor released version. It will then be easier to deploy 'FastPG' as a dependency in the development of other Bioconductor packages.

Thanks,
Tushar

[sararselitsky]

Thanks! We have no plans to turn this into a bioconductor package at this time, but I can check with the rest of the group.

…

On Thu, Aug 18, 2022 at 2:49 PM Tusharkanti Ghosh ***@***.***> wrote: Hi, Congratulations for writing such a wonderful compact package. I was wondering if this package will be available on Bioconductor released version. It will then be easier to deploy 'FastPG' as a dependency in the development of other Bioconductor packages. Thanks, Tushar — Reply to this email directly, view it on GitHub <#21>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACQD6T2NKJYYXV7GH2M2BUDVZ2AVJANCNFSM566FOJMQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[jefferys]

The included C++ code for Grappolo makes it difficult to create a package that meets Bioconductor standards. It is essentially just all the code for an independent application wrapped into a package with minimal tweaking. It does a lot of stuff unfriendly to R including ignoring R's random number system, R's error handling and interrupts, and it writes directly to the console. We could remove Grappolo and make it an externally installed dependency, but that would make FastPG harder to set up and use, not easier! Alternately, we could more fully integrate Grappolo into R, making FastPG more robust in the process, but that would take significant development effort.

[leeleavitt]

Any chance going the route of publishing on bioconda?

[sararselitsky]

No plans for it. What would it entail?

…

On Tue, Jan 16, 2024 at 6:43 PM Lee S. Leavitt ***@***.***> wrote: Any chance going the route of publishing on bioconda? — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACQD6TZ4G3DIVU7KBVPWAWLYO4GAHAVCNFSM566FOJM2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOBZGQ3DSNBZGI2Q> . You are receiving this because you commented.Message ID: ***@***.***>

[leeleavitt]

See here for more information. Specifically, refer to this section.

Publishing to bioconda offers numerous benefits, with the primary one being installation reproducibility. The packages are distributed as binaries, which results in exceptionally fast installations.

A significant issue with R is the absence of an environment for installing all dependencies. This can lead to modifications in the global system to accommodate the software, potentially leaving the state of other software uncertain.

My objective was to install your software into a Conda environment, but I've encountered failure in approximately five attempts.