GNPS analysis doesn't work

[joewandy]

A user reported that her GNPS analysis failed to run on PALS Viewer.

Thanks for offering to take a look at our issue!
I'm attaching the metadata here. This is the GNPS link: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=d268caea906a4383873c156c4e54e411

Metadata is attached:
new_updated_influenza_metadata_PALS_1.zip

[justinjjvanderhooft]

Note that that GNPS run is in release 21, which is quite some versions back! That could be a reason if some file structures have changed since then....

[mccall-lab-OU]

I tried cloning the job to the newest GNPS version (28.2), and unfortunately that didn't fix the problem.
I also downloaded the metadata straight from the GNPS job and re-formatted it to match your example (see attached).
We still get the error as follows.
GNPS job link is: https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=0c369ab19a6f44798180f123da5f9134
Thanks!

Error text:
None of the sample names in the metadata ['pos_lung_tissue_A1_P54_1_1_infected.mzXML' 'pos_lung_tissue_A2_P54_1_2_infected.mzXML' 'pos_lung_tissue_A3_P54_1_3_infected.mzXML' 'pos_lung_tissue_A4_P54_1_4_infected.mzXML' 'pos_lung_tissue_A5_P54_1_5_infected.mzXML' 'pos_lung_tissue_A6_P54_1_6_infected.mzXML' 'pos_lung_tissue_A7_P54_1_7_infected.mzXML' 'pos_lung_tissue_A8_P54_1_8_infected.mzXML' 'pos_lung_tissue_A9_P54_1_9_infected.mzXML' 'pos_lung_tissue_A10_P54_1_10_infected.mzXML' 'pos_lung_tissue_A11_P54_1_11_infected.mzXML' 'pos_lung_tissue_A12_P54_1_12_infected.mzXML' 'pos_lung_tissue_B1_P54_2_1_infected.mzXML' 'pos_lung_tissue_B2_P54_2_2_infected.mzXML' 'pos_lung_tissue_B3_P54_2_3_infected.mzXML' 'pos_lung_tissue_B4_P54_2_4_infected.mzXML' 'pos_lung_tissue_B5_P54_2_5_infected.mzXML' 'pos_lung_tissue_B6_P54_2_6_infected.mzXML' 'pos_lung_tissue_B7_P54_2_7_infected.mzXML' 'pos_lung_tissue_B8_P54_2_8_infected.mzXML' 'pos_lung_tissue_B9_P54_2_9_infected.mzXML' 'pos_lung_tissue_B10_P54_2_10_infected.mzXML' 'pos_lung_tissue_B11_P54_2_11_infected.mzXML' 'pos_lung_tissue_B12_P54_2_12_infected.mzXML' 'pos_lung_tissue_C1_P54_3_1_infected.mzXML' 'pos_lung_tissue_C2_P54_3_2_infected.mzXML' 'pos_lung_tissue_C3_P54_3_3_infected.mzXML' 'pos_lung_tissue_C4_P54_3_4_infected.mzXML' 'pos_lung_tissue_C5_P54_3_5_infected.mzXML' 'pos_lung_tissue_C6_P54_3_6_infected.mzXML' 'pos_lung_tissue_C7_P54_3_7_infected.mzXML' 'pos_lung_tissue_C8_P54_3_8_infected.mzXML' 'pos_lung_tissue_C9_P54_3_9_infected.mzXML' 'pos_lung_tissue_C10_P54_3_10_infected.mzXML' 'pos_lung_tissue_C11_P54_3_11_infected.mzXML' 'pos_lung_tissue_C12_P54_3_12_infected.mzXML' 'pos_lung_tissue_D1_P54_4_1_uninfected.mzXML' 'pos_lung_tissue_D2_P54_4_2_uninfected.mzXML' 'pos_lung_tissue_D3_P54_4_3_uninfected.mzXML' 'pos_lung_tissue_D4_P54_4_4_uninfected.mzXML' 'pos_lung_tissue_D5_P54_4_5_uninfected.mzXML' 'pos_lung_tissue_D6_P54_4_6_uninfected.mzXML' 'pos_lung_tissue_D7_P54_4_7_uninfected.mzXML' 'pos_lung_tissue_D8_P54_4_8_uninfected.mzXML' 'pos_lung_tissue_D9_P54_4_9_uninfected.mzXML' 'pos_lung_tissue_D10_P54_4_10_uninfected.mzXML' 'pos_lung_tissue_D11_P54_4_11_uninfected.mzXML' 'pos_lung_tissue_D12_P54_4_12_uninfected.mzXML' 'pos_lung_tissue_E1_P54_5_12_uninfected.mzXML' 'pos_lung_tissue_E2_P54_5_11_uninfected.mzXML' 'pos_lung_tissue_E3_P54_5_10_uninfected.mzXML' 'pos_lung_tissue_E4_P54_5_9_uninfected.mzXML' 'pos_lung_tissue_E5_P54_5_8_uninfected.mzXML' 'pos_lung_tissue_E6_P54_5_7_uninfected.mzXML' 'pos_lung_tissue_E7_P54_5_6_uninfected.mzXML' 'pos_lung_tissue_E8_P54_5_5_uninfected.mzXML' 'pos_lung_tissue_E9_P54_5_4_uninfected.mzXML' 'pos_lung_tissue_E10_P54_5_3_uninfected.mzXML'] matches the columns names in the measurements []. Please check the names again.

metadata_influenza_fromGNPS.csv

[justinjjvanderhooft]

@joewandy this may be due to a restructuring at the GNPS end in newer versions, could you check what happens in the PALS code after downloading the GNPS results? The metadata file loads fine in GNPS, so that means that the file names should be correct: https://gnps.ucsd.edu/ProteoSAFe/result.jsp?task=0c369ab19a6f44798180f123da5f9134&view=view_metadata - perhaps the folder structure changed a bit and PALS can no longer find the relevant file to match the names to?