search

meowcat / MSnio

3 stars 3 forks source link

Package name #3

Open lgatto opened 5 years ago

lgatto commented 5 years ago

I just had the MSnbase devel call with @jorainer and he updated me about the discussions at the recent metabolomics workshop, including the MS scheme to harmonise the nomenclature and schemes in different metabolomics MS2 databases.

My understanding is that this package is meant to implement this MS scheme at the R level. If so, I would suggest to change the package name to reflect this goal more specifically, possibly naming it MSschemer (or whatever capitalisation seems more appropriate). This would also avoid confusion with other IO concepts/developments related to MS data.

Thank you in advance for considering my suggestion.

meowcat commented 5 years ago

Hi,

The name is a working title and can change. "MSnio" was chosen for now because a large practical part of the package will be canonical, metadata-rich high-quality import/export of files such as msp, mgf, MassBank. How this lines up with the Spectra plans I don't know and should be part of our discussions.

meowcat commented 5 years ago

@michaelwitting @Treutler What do you think of MSnPort? ("port" for portable (between formats), import and export.) I agree with @lgatto that the name MSnio promises something different from what we do (that would be more an appropriate name for mzR, for example.)

@lgatto do you take issue with using "MSn" which might allude to MSnbase? I just find the name more harmonic like this.

Treutler commented 5 years ago

If all are fine with the prefix "MSn" (because MSnbase is the basis) then I find names like MSnConvert, MSnPort, and MSnAdapter suitable. IMHO MSschemer is a bit unintuitive and MSnio is a bit unspecific.

lgatto commented 5 years ago

My suggestion to use MSschemer was because, as far as I understand, it is an implementation of the MS scheme to translate between schemes and nomenclature of different (metabolomics) spectral library databases. I don't mind to use MSn to make the link with MSnbase explicit (I think it's a good idea), as it is likely we will make use of your effort for some of our backends. So MSnScheme (with or without r) would also be a fair refection of the nature of the package. But I'll leave it to the stakeholders.

Following up from @Treutler's suggestions, maybe MSnDbAdapter?

michaelwitting commented 5 years ago

If would prefer MSnSchemer or MSnScheme. MSnDbAdapter might be also okay, but I wanted to have also query spectra formats in their, e.g. for MetFrag, Sirius etc... The Db might be missleading here.

meowcat commented 5 years ago

Then possibly MSnSchema since "Schema" is more commonly used compared to "scheme" in databases. But I am still unsure this is good, because it leaves the parser / writer aspect out.

sneumann commented 5 years ago

MSnbRidge, alluding to http://bioconductor.org/packages/3.9/bioc/html/BridgeDbR.html ? Ok, MSnBridge as I had difficulties finding the Bride in the name myself. Yours, Steffen