proteinGroups.txt and other files were empty when using Maxquant running 84 samples on linux

[lishr0007-spec]

Describe the bug
proteinGroups.txt, Oxidation (M)Sites.txt, matchedFeatures.txt, matchedFeatures.txt, modificationSpecificPeptides.txt were empty while peptides.txt, evidence.txt and other files are normal when using Maxquant running a 84-sample search.

data/MaxQuant_output/msms.txt 17533124 KB
data/MaxQuant_output/evidence.txt 1118928 KB
data/MaxQuant_output/peptides.txt 310712 KB
data/MaxQuant_output/a-2102903091/peptides 261624 KB
data/MaxQuant_output/a604834882/peptides 261624 KB
data/MaxQuant_output/a-1120265650/modPeptides0 232208 KB
data/MaxQuant_output/a519827151/modPeptides0 135376 KB
data/MaxQuant_output/a-1291871554/modPeptides0 135244 KB
data/MaxQuant_output/a-1219179865/peptides 131876 KB
data/MaxQuant_output/a-755779314/peptides 131876 KB
data/MaxQuant_output/a1001199047/modPeptides0 102236 KB
data/MaxQuant_output/a-1120265650/modPeptides.ind 92888 KB
data/MaxQuant_output/a-2102903091/peptides.ind 85748 KB
data/MaxQuant_output/a604834882/peptides.ind 85748 KB
data/MaxQuant_output/a506284000/modPeptides0 63528 KB
data/MaxQuant_output/a-651136641/modPeptides0 63436 KB
data/MaxQuant_output/a519827151/modPeptides.ind 54152 KB
data/MaxQuant_output/a-1291871554/modPeptides.ind 54100 KB
data/MaxQuant_output/a-1219179865/peptides.ind 43196 KB
data/MaxQuant_output/a-755779314/peptides.ind 43196 KB
data/MaxQuant_output/a1001199047/modPeptides.ind 40896 KB

-rw-rw-r--+ 1 lsr lsr 0 Sep 2 19:58 'Oxidation (M)Sites.txt'
-rw-rw-r--+ 1 lsr lsr 4.3K Sep 2 19:58 parameters.txt
-rw-rw-r--+ 1 lsr lsr 304M Sep 2 20:07 peptides.txt
-rw-rw-r--+ 1 lsr lsr 0 Sep 2 19:58 proteinGroups.txt
-rw-rw-r--+ 1 lsr lsr 172K Sep 2 10:55 sdrf.tsv
-rw-rw-r--+ 1 lsr lsr 69K Sep 2 20:05 summary.txt
-rw-rw-r--+ 1 lsr lsr 257K Sep 2 19:58 tables.pdf
-rw-rw-r--+ 1 lsr lsr 1.1G Sep 2 20:03 evidence.txt
-rw-rw-r--+ 1 lsr lsr 0 Sep 2 19:58 matchedFeatures.txt
-rw-rw-r--+ 1 lsr lsr 0 Sep 2 19:58 modificationSpecificPeptides.txt
-rw-rw-r--+ 1 lsr lsr 70K Sep 2 19:58 mqpar.xml
-rw-rw-r--+ 1 lsr lsr 3 Sep 2 11:13 ver

[WalterViegener]

what version did you use?

[lishr0007-spec]

what version did you use?

Thank you very much for replying! I used MaxQuant_v2.8.1.0, the dotnet version is 8.0.424.

[WalterViegener]

ok, please upload your mqpar file here

[lishr0007-spec]

ok, please upload your mqpar file here

The xml file has been loaded. Thank you !

[WalterViegener]

Some more info I need:

1. How big is your RAM ?
2. did it write ...error.txt files in the combined/proc folder?
3. please upload also the combined/txt/#runningTimes.txt file

[lishr0007-spec]

Some more info I need:

1. How big is your RAM ?
2. did it write ...error.txt files in the combined/proc folder?
3. please upload also the combined/txt/#runningTimes.txt file

1. The RAM is "MemTotal: 528006104 kB".

2 and 3. There are no error.txt files in the combined/proc folder and there is no txt folder generated under combined folder. The #runningTimes.txt and Calculating_PEP 11.finished.txt were the only files deposited in the combined/proc folder and both of them were uploaded. The evidence.txt, peptides.txt, msms.txt and proteinGroups.txt files are in the output folder I specified.

[WalterViegener]

Thanks — that helps, but those two files are from your new run (started Sep 7 21:28), not from the failed Sep 2 one, so unfortunately they can't tell us much about the empty tables:

• MaxQuant deletes the whole combined/proc folder at the start of every run. So the new run erased anything the Sep 2 run had left there, including any .error.txt. That's why you only see #runningTimes.txt and one .finished.txt.
• My question 3 was misleading, sorry: you set a custom output folder, so the run's #runningTimes.txt is copied to MaxQuant_output/, not combined/txt/.
• Also, the new run seems to use numThreads = 4, while the mqpar you uploaded says 32 — so it may be a different parameter file.

One thing I can already confirm: matchedFeatures.txt being empty is expected in your setup, because "match unidentified features" is turned off. That file is only filled when that option is on, so it's not part of the problem.

For the other three (proteinGroups.txt, Oxidation (M)Sites.txt, modificationSpecificPeptides.txt), the sizes tell us the result wasn't actually empty — peptides.txt (304 MB) can only be written if the protein groups were built correctly. So those three tables failed while being written, and MaxQuant currently doesn't report that failure clearly. I'd like to find out why they failed.

Could you send, from the Sep 2 run:

1. The console output (stdout/stderr) of that run, especially the last ~50 lines. Did it print anything like MaxQuant stopped with error code ..., or did it appear to finish normally?
2. MaxQuant_output/#runningTimes.txt, if it exists with a Sep 2 date.
3. MaxQuant_output/mqpar.xml (the copy MaxQuant wrote at 19:58 on Sep 2) — that's the parameter file actually used for the failed run.
4. The output of:
   dmesg -T | grep -iE 'killed process|out of memory'
   filtered around Sep 2, 19:58–20:07. This tells us whether the Linux kernel killed the processes (e.g. out of memory).

And one question: is the Sep 7 run still going, or did it stop after "Calculating PEP"? Its log stops right there.

Note that rerunning into the same combined folder overwrites the previous run's data, so if you still have the Sep 2 folder somewhere untouched, please don't delete it.

As a workaround in the meantime, you could try lowering the number of threads.