MSstats Big and selection of intensity column

[Roneldine Mesidor]

Hi,

I was just wondering how MSstatsBig selects the "intensity" column from the MSstats Report. 

For the regular SpectronauttoMSstatsFormat ccommand, I could specify which column I want to use for the MSstats processing steps. However, MSstatsBig doesn't allow for that anymore. Is there a default selection? If so, which one?

Thank you,

Roneldine Mesdor

[Anthony Wu]

Hi,

For the bigSpectronauttoMSstatsFormat function, you could set the intensity parameter to specify the column you want for processing.  The default is "F.NormalizedPeakArea", but you can also use "F.PeakArea" too.  

Example:

converted_data <- bigSpectronauttoMSstatsFormat(
    system.file("extdata", "spectronaut_input.csv", package = "MSstatsBig"),

    "output_file.csv",
    backend="arrow"

    intensity = "F.PeakArea")
converted_data <- dplyr::collect(converted_data)
head(converted_data)

Let me know if you have any problems with that.

Thanks,

Tony

[Roneldine Mesidor]

Hi,

Thanks for the answer. I have already tried to use the F.PeakArea, but I got the following error: 

Error in bigSpectronauttoMSstatsFormat("4_StudyIV_FreshvsCryo_Passaging/1_WholeCell_Data/1_Data/Raw/336631_MSstatsReport_StudyIV.tsv", : unused argument (intensity = "F.PeakArea")

If I can't select a specific column for processing, should I just do logTrans and not Normalization for the dataProcess function? I thought it was best to do log2 transformation before normalization. I may be wrong so please let me know what the best approach is moving forward.

Thank you,

Roneldine

[Anthony Wu]

Hi,

Could you try installing MSstatsBig with this command:

devtools::install_github("Vitek-Lab/MSstatsBig", build_vignettes = TRUE)

Upon further investigation, it looks like we did not push our most recent changes adding the `intensity` parameter to Bioconductor.  We will push to Bioconductor asap.

In the meantime, installing the code from Github should unblock you.

Thanks,

Tony