Version above 1.2.7 causes issue with groupComparisonTMT function

[Trent Hinkle]

Hi, 

I am a heavy user of MSstatsTMT and have recently upgraded from version 1.2.7 to version 1.5.6 (Also 1.4.6).

Upon upgrading versions (Up from 1.2.7) I have experienced an issue with MSstatsTMT function groupComparisonTMT. Within the output of groupComparisonTMT I get no Pvalues or Log2FC values and in the "issue" column for all of my proteins I get "unfittableModel". However, using the older version of MSstatsTMT (Version 1.2.7) I am able to get proper statistics (Pvalues and Log2FC values, etc).

To this email I am attaching the input file to groupComparisonTMT as well as the output I get from that function for version 1.2.7 as well as version 1.5.6.

I was wondering if someone could shed some light on why this is happening in the newer versions of MSstatsTMT?

Any help would be greatly appreciated.

Thank you,

Trent Hinkle

[Krystine Mansfield]

Was this issue ever resolved? I am seeing the same thing in the new release 2.0

[thuan...@gmail.com]

Yes, I believe the issue has been fixed in the previous release. The error you saw may be due to other reasons. If possible, could you please share the input you used to groupComparisonTMT function?

-Ting

[Krystine Mansfield]

Actually, I was just using the example data that comes with the package. I decided to use the "Median" method for protein summarization as that is what I had used previously used with MSstats on LFQ data and was familiar with the calculation. As you can see in the attached screenshot, there are no stats values and the issue says "unfittableModel". 

library(MSstatsTMT)

quant <- PDtoMSstatsTMTFormat(input = raw.pd, annotation = annotation.pd, which.proteinid = 'Protein.Accessions')

quant.median <- proteinSummarization(quant, method="Median", global_norm=TRUE, reference_norm=TRUE, remove_norm_channel = TRUE, remove_empty_channel = TRUE)

comparison=matrix(c(-1,0,0,1),nrow=1)

row.names(comparison)="1-0.125"

colnames(comparison)= c("0.125", "0.5", "0.667", "1")

test.contrast = groupComparisonTMT(data = quant.median,contrast.matrix = comparison, moderated = TRUE)

[thuan...@gmail.com]

I run the example code on my laptop and the output seems no problem. 

Which version of MSstatsTMT do you use? Could you run the following code to check the package version info? 

sessionInfo()

-Ting

[Krystine Mansfield]

I thought I had version 2.0 but since I am running R3.6.3, MSstatsTMT v1.4.6 was installed. Is this too far back?

[thuan...@gmail.com]

Yes, you probably need to install R 4.X.X (https://cloud.r-project.org/) and MSstatsTMT 2.0.0. This issue should be fixed in the latest version of MSstatsTMT

[Krystine Mansfield]

Thank you. That fixed the issue.