Nesvilab/FragPipe

Fragpipe 22 doesn't generate MSStats csv output file

Closed this issue · 8 comments

bnewt2k commented

Hi,
Running DIANN w Fragpipe
Fragpipe 22 doesn't generate msstats.csv even though box is checked.
Fragpipe 21.1 did write csv.

Thanks
Bill

log_2024-06-05_10-05-51.txt

- Upload your log file
(If a log file hasn't been generated, go to the 'Run' tab in FragPipe, click 'Export Log', zip the resulting "log_[date_time].txt" file to avoid truncation, then attach the zipped file by drag & drop here.)

- Describe the issue or question:

fcyu commented

I think you need to specify the experiment and bioreplicate groups in the workflow tab enable msstats.csv, because MSstats needs the experiment structure.

Best,

Fengchao

bnewt2k commented
fcyu commented

No, you didn't.

image

bnewt2k commented
bnewt2k commented

Hi Fengchao,

Just tried again. No msstats.csv

log_2024-06-05_15-02-56.txt
FP_22_Workflow
FP_22_QuantDIA

Thanks,
Billy

fcyu commented

Hi Billy,

I think the files are messed up because

  1. Although you specified the experiment and bioreplicate groups, they were not well recorded when running the jobs:
LCMS files:
  Experiment/Group: 
  (if "spectral library generation" is enabled, all files will be analyzed together)
  - C:\Champions_10AML_MZML\Ana\20230728_SJMS_Eclipse_FAIMS_Champ_Reg_DIA_Ana_2of20ul_Sample_01.mzML	DIA
  - C:\Champions_10AML_MZML\Ana\20230728_SJMS_Eclipse_FAIMS_Champ_Reg_DIA_Ana_2of20ul_Sample_02.mzML	DIA
  - C:\Champions_10AML_MZML\Ana\20230728_SJMS_Eclipse_FAIMS_Champ_Reg_DIA_Ana_2of20ul_Sample_03.mzML	DIA
  - C:\Champions_10AML_MZML\Ana\20230728_SJMS_Eclipse_FAIMS_Champ_Reg_DIA_Ana_2of20ul_Sample_04.mzML	DIA
  - C:\Champions_10AML_MZML\Ana\20230728_SJMS_Eclipse_FAIMS_Champ_Reg_DIA_Ana_2of20ul_Sample_05.mzML	DIA
  - C:\Champions_10AML_MZML\Ana\20230728_SJMS_Eclipse_FAIMS_Champ_Reg_DIA_Ana_2of20ul_Sample_06.mzML	DIA
  1. You did not enable the spec lib generation step, but DIA-NN quant still worked. It implies that there are old files (C:\May7_Test2\Old_Library\library_Hum_Panc_Ecl.tsv) from the previous job.....
    Cmd: [START], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [CheckCentroid], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [WorkspaceCleanInit], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [MSFragger], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [MSBooster], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [Percolator], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [ProteinProphet], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [PhilosopherDbAnnotate], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [PhilosopherFilter], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [PhilosopherReport], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [WorkspaceClean], Work dir: [C:\Jun05_frag22_test_simple]
    Cmd: [DIA-NN], Work dir: [C:\Jun05_frag22_test_simple]

...

speclibgen.run-speclibgen=false

...

[0:00] Loading spectral library C:\May7_Test2\Old_Library\library_Hum_Panc_Ecl.tsv
[0:05] Finding proteotypic peptides (assuming that the list of UniProt ids provided for each peptide is complete)
[0:05] Spectral library loaded: 9693 protein isoforms, 9693 protein groups and 205632 precursors in 178396 elution groups.
[0:05] Encoding peptides for spectra and RTs prediction
[0:05] Predicting spectra and IMs
[0:43] Decoding predicted spectra and IMs
[0:46] Saving the library to lib.predicted.speclib
[0:46] Initialising library
...

Could you

  1. Delete all files related to FragPipe
  2. Download FragPipe, and other tools to a new folder (DO NOT put them to a folder having any files).
  3. Load the built-in workflow, your LC-MS files
  4. Specify your fasta file
  5. Go to the run tab, specify the output folder to a new one, and run

Thanks,

Fengchao

bnewt2k commented

Thank You,
Fengchao

I still need guidance on the library usage, and generation or not..?
I can still reuse the old library right?
(I would rather not rebuild library from 30+ DDA runs); also won't that inhibit the grouping just showed

Or do I have to generate a library from every time I want to export a msstats.csv ?

I will put the old library into the new folder without any other files.

Thanks
Billy

bnewt2k commented

*** Update ***

Hi All,

So I did get FP_22 to output a msstats.csv file, after trying some things.

I deleted all fragpipe versions and tools I had been using from folders on C drive. ( although I did NOT go into c:\windows\users\Appdata\ etc.. and delete anything)
I reinstalled FP 22_jre and all tools including python.
I did not reinstall java or C, I thought those should be fine.

log_2024-06-06_14-12-28.txt

I did add experiment and bioreplicate values in workflow tab.
I still tried using the existing library created at least two FP version ago ~May 2023.
That still failed to produce a msstats.csv.

Then I tried to not use the existing spectral library I had been using.
I added 8 DDA files I had used to make that library.
I selected to generate a new library file.
I used a new download of uniprot human reviewed fasta file.
Then FP wrote msstats.csv.

I suspect that the old library caused most problems...
AND / OR that selecting generate new library fixed them.

I will test with new library alone minus DDA files to confirm. It would be common to use an existing library without building a new one each time.

Thanks all
Billy