Can't assign 4 names to a 0 column data.table

bw li

unread,

Apr 8, 2022, 2:19:11 AM4/8/22

to MSstats

I want to use MSstats to analyze Maxquant results, I have some trouble using MaxQtoMSstatsFormat, the following is the error：

INFO [2022-04-08 14:14:32] ** Features with one or two measurements across runs are removed. INFO [2022-04-08 14:14:32] ** Run annotation merged with quantification data. INFO [2022-04-08 14:14:32] ** Multiple fractionations exist: 6 fractionations per MS replicate. INFO [2022-04-08 14:14:42] ** Features with one or two measurements across runs are removed. INFO [2022-04-08 14:14:42] ** Fractionation handled. Error in setnames(x, value) : Can't assign 4 names to a 0 column data.table

I looked at similar questions but still didn't solve the problem, here is my code：

proteinGroups <- data.table::fread("~/biodata/CNHPP_HCC/L006/combined/txt/proteinGroups.txt")
infile<-data.table::fread('~/biodata/CNHPP_HCC/L006/combined/txt/evidence.txt')
infile$Intensity <- as.numeric(infile$Intensity)
raw <- MaxQtoMSstatsFormat(evidence=infile, annotation=annot,proteinGroups=proteinGroups)

How can I fix this problem?

Thanks,

LeeLee

Mateusz Staniak

unread,

Apr 12, 2022, 10:39:16 AM4/12/22

to MSstats

Hi,

can I see a small example of data where the function fails?

Kind regards,

Mateusz

bw li

unread,

Apr 12, 2022, 11:04:32 PM4/12/22

to MSstats

annot.csv

evidence_part.txt

proteinGroups_part.txt

bw li

unread,

Apr 12, 2022, 11:05:41 PM4/12/22

to MSstats

Hi, Mateusz

I uploaded part of the file, please take a look.

Thanks,

LeeLee

在2022年4月13日星期三 UTC+8 11:04:32<bw li> 写道：

Mateusz Staniak

unread,

Apr 15, 2022, 3:14:33 PM4/15/22

to MSstats

Hi,

I looked at the example. Could you please check your evidence to see if there are any features (= combinations of (protein name, peptide sequence, precursor charge)) with at least 3 non-missing observations across runs? In the example, the error is due to the fact that we remove features with less than 3 observations, and after this filtering there are no observations left.

Kind regards,

Mateusz

bw li

unread,

Apr 17, 2022, 9:41:54 PM4/17/22

to MSstats

Hi， Mateusz

I don't quite understand. If I understand you correctly, do you mean that this kind of problem occurs when the protein name, peptide sequence, and precursor charge are missing at the same time? But I didn't observe this happening.

LeeLee

Carlos Gonzalez

unread,

Jun 26, 2025, 6:54:36 PMJun 26

to MSstats

Sorry to bring this up from the grave, but I am seeing this error as well and didn't want to start a new thread. From what I can ascertain from the reply above, you need at least 3 observations of a feature otherwise it removes it correct? This is probably a corner case but occasionally some people just run two files and it seems as though under these conditions, nothing would pass the filter, correct? I have attached a snippet to peruse.

Cheers

Carlos

msstats_small.csv

Mateusz Staniak

unread,

Jun 27, 2025, 4:49:55 AMJun 27

to MSstats

Hi,

can I see an annotation file, too? Just so I can run the converter and reproduce the issue on my end.

Apologies for a lack of reply to an earlier question. I meant features such that intensities are NA.

Kind regards,
Mateusz

Carlos Gonzalez

unread,

Jun 27, 2025, 10:18:20 AMJun 27

to MSstats

Hi Mateusz,

Apologies my situation is not exactly like the issue above. For this, there is no annotation file as it was straight out of FragPipe. I can give you the whole file if you like (just let me know where to upload), or any other file from the FragPipe output.