Question on SpectronauttoMSstatsFormat

[Yunran]

Hi,

This is a great tool but I have hard time implement it.

I am wondering how MSstats generates SUBJECT column in the dataProcess()$ProteinLevelData. Does it comes from R.Replicate in raw Spectronaut output? Or does it come from the annotation file by setting SpectronauttoMSstatsFormat(anno = xx)?

If it does come from R.Replicate, how should I specify the R.Replicate?

Suppose I have 3 condition, 5 donors for each condition, 2 tech replicates for each donor. In total 3 x 5 x 2 runs. In the R.Replicate, should I encode using C1D1T1 ? 

I am asking this question is because, I have large raw output from Spectonaut so that I cannot apply SpectronauttoMSstatsFormat directly. Instead, I can only use "MSstatsBig" R package to replace function SpectronauttoMSstatsFormat. 

Thank you so much for your time and help.

Best,

Yunran

[Anthony Wu]

Hi,

Both ways can work: you can give the bioreplicate ID using R.Replicate in the raw data, or you can provide it in the annotation file when using SpectronauttoMSstatsFormat(anno = ...).

In your case, the bioreplicate ID should represent a unique donor (for example, C1D1 for condition 1, donor 1). You do not need to add the technical replicate number (so not C1D1T1).  If two runs have the same bioreplicate ID and condition, MSstats will automatically understand that they are technical replicates.

For more details and an example, you can look at Supplementary Table 1 in the MSstats v4 paper (link).  See Supplementary Table 2 in case your experiment is a repeated measures / time-course design.

Thanks,

Tony

[Anthony Wu]

Posting a follow up question in case others have similar questions:

"If I have donor 1-5 across 2 conditions (each donor have two conditions), should I use "D1" instead of "C1D1", correct?"

My response: Here, I would use "D1" instead of "C1D1".  Supplementary table 3 & supplementary figure 6 should be good references (link).  This study can be treated like a paired design.