search

sneumann / xcms

This is the git repository matching the Bioconductor package xcms: LC/MS and GC/MS Data Analysis
Other
183 stars 80 forks source link

featureSpectra with Biocparallel #645

Open 13479776 opened 1 year ago

13479776 commented 1 year ago

Hi guys,

I used featureSpectra function in XCMS, and found it was so slow, i request if it is possible to use Biocparallel package for processing the sampels with featureSpectra function?

Thanks

Best wishes, Hees

jorainer commented 1 year ago

Dear Hees,

can you please specify how you called featureSpectra (i.e. using which parameters)? Depending on what object should be returned the call can be slower or faster.

13479776 commented 1 year ago

Hi jorainer,

I just want to make the featureSpectra run with Biocparallel, i have edit it and for each feature with Biocparallel. Thanks

Best wishes, Hees

jorainer commented 1 year ago

Dear Hees,

I am currently working on a new version of xcms that uses the Spectra and MsExperiment R packages instead of MSnbase. Extracting spectra for chromatographic peaks (and then also features) will be much faster because there will be no need to convert between different types of objects.

The code base is currently in the spectra branch.

mscbmlab commented 1 year ago

Hi jorainer,

It is amazing, Thanks, i will try the new Spectra.

Best wishes, Hees

jorainer commented 1 year ago

You can try it if you want, but it's not yet done (specifically featureSpectra is not yet implemented for the new classes.

Also, you would need to import the data in a different way:

data <- readMsExperiment(files, pd)

Where files is the character vector with the files to import and pd a data.frame with the sample annotations (one row per sample/file - rows need to be in the same order than files). I did not yet update the vignette on the new way to import data because we're not yet sure how to best do that...