Datasets from SIM-XL, Mascot and ProteomeDiscover in PRIDE

[ypriverol]

We have to find out the list of datasets with the following conditions:

• mzIdentml 1.2 (Complete submission)
• Peak list files corresponding to the mzid
• Software that produces the files ProteomeDiscover / Mascot / SIM-XL
• Crosslinking (this is the most important one )

Please lets update the list in this issue.

[sureshhewabi]

Datasets in PRIDE with "crosslink" or "cross-link" word in TITLE which contains mzIdentML files:

Ordered by priority:

• PXD029252, Proteome Discoverer - Failing - No mgf presents
• PXD021417, TOPP software - Datasets from SIM-XL, Mascot and ProteomeDiscover in PRIDE #63 (comment)
• PXD014359,TOPP software - Datasets from SIM-XL, Mascot and ProteomeDiscover in PRIDE #63 (comment)
• PXD014520,TOPP software - spectra ref to Raw files
• PXD014523,TOPP software - Spectra Location Unknown
• PXD026603, ProteomeDiscover - Datasets from SIM-XL, Mascot and ProteomeDiscover in PRIDE #63 (comment)
• PXD028039, xiFDR software: Failing no mgf
• PXD050457, MassIVE dataset, looks like it works in xiView
• PXD018935,Proteome Discoverer
• PXD012759,Proteome Discoverer

Need to check version 1.2, the corresponding peak list and producer

[colin-combe]

as noted in meeting, they might not be complete submissions
(what does the strikeout represent above? PXD018935 / PXD012759)

[colin-combe]

"crosslink" or "cross-link" word in TITLE

wasn't there a "crosslink" tag people were referring to? (I don't know but people spoke of this)

[ypriverol]

We will continue with different combinations. We will et you know when errors start to happen.

[colin-combe]

OK, great, thanks!

[ypriverol]

@sureshhewabi reported the following error in this one:

PXD014359 - Error parsing C_Lee_141014_CRM_dialysis_NCE20_2.mzid
MzIdParseException ('XMLSyntaxError', ("Start tag expected, '<' not found, line 1, column 1",))
2024-04-04 09:35:10 - main - ERROR - ('XMLSyntaxError', ("Start tag expected, '<' not found, line 1, column 1",))

[colin-combe]

i guess the error message is correct and it is not valid XML

[ypriverol]

Another similar error for OpenMS

Error parsing XLpeplib_Beveridge_QEx-HFX_DSS_R1.mzid
MzIdParseException ('XMLSyntaxError', ("Start tag expected, '<' not found, line 1, column 1",))
2024-04-04 14:28:13 - parser.process_dataset - ERROR - ('XMLSyntaxError', ("Start tag expected, '<' not found, line 1, column 1",))

[sureshhewabi]

Schema seems valid in XLpeplib_Beveridge_QEx-HFX_DSS_R1.mzid file from PXD021417:
xmllint --noout --schema mzIdentML1.2.0.xsd XLpeplib_Beveridge_QEx-HFX_DSS_R1.mzid > XLpeplib_Beveridge_QEx-HFX_DSS_R1_output_file1.txt 2>&1

There should be an issue with the parser. @colin-combe any idea?

[colin-combe]

yes, could be a issue with parser. Or perhaps something to do with character encoding. I looked into and was confused.

@sureshhewabi - I'm not sure what your xmllint command does ?

[sureshhewabi]

It is a command to check the schema validity against the schema definition file(xsd)

[colin-combe]

one problem is the empty location attribute for spectra data:

XLpeplib_Beveridge_QEx-HFX_DSS_R3.mzid, line 527672:
<SpectraData location="" id="SDAT_1534307058980521776">

It is a required attribute, but empty string is enough to make the file schema valid (http://www.datypic.com/sc/xsd/t-xsd_anyURI.html).

But this isn't the only problem, there's something else that's still mysterious...

[sureshhewabi]

This means we cannot use this dataset for us anyway, isn't it? because we cannot find the peaklist file

[colin-combe]

we could manually fix the location.
But also, yes, there is a problem with the parser. It is requiring some elements that are optional. (Breaks if they're not there.)
I'll provide an update (will make PR when its fixed).
I would stop testing datasets until this is fixed.

[colin-combe]

think this fixes a problem - #64
the datset with the empty location still won't work, but maybe some of the other ones throwing errors like that will.

sorry about that

[sureshhewabi]

PXD021417 Dataset Issues:

• Spectra Data Location was not available - Manually inserted as we can see one to one relationship between mzidentml files and mzML files
• Issues with Score
• Run Name is missing
• <Seq> is missing

[sureshhewabi]

PXD026603 Dataset Issues:

• ModificationParam

parser.process_dataset - INFO - parsing AnalysisProtocolCollection- start
Error parsing GPR158-RGS7-Gb5_CONSENSUS.mzid
KeyError 'ModificationParams'
parser.process_dataset - ERROR - 'ModificationParams'

• No peaklist file found
  Datasets from SIM-XL, Mascot and ProteomeDiscover in PRIDE #63 (comment)

[colin-combe]

thanks, will check it

[colin-combe]

similar to before - parser was treating things that are optional as if they were required
fixed by #65

re PXD026603 - the peaklists are missing?

[sureshhewabi]

Yes, peakfile is missing too:
<SpectraData location="C:\Users\griffinlab.PG18844\Dropbox (Scripps Research)\Griffin Lab\fusion lumos\TSS\XLMS\20210129 GPR158 Complex\GPR158_Complex_XLMS\GPR158-RGS7-Gb5_CONSENSUS.mzML" name="MzML spectra file" id="ID_MZML_FILE_with_spectra"> but GPR158-RGS7-Gb5_CONSENSUS.mzML is not available

[colin-combe]

We have to find out the list of datasets with the following conditions:

• mzIdentml 1.2 (Complete submission)

would these actually have complete submission status? I thought complete submission status wasn't previously being given to crosslinking data?

[colin-combe]

PXD021417 Dataset Issues:

• Spectra Data Location was not available - Manually inserted as we can see one to one relationship between mzidentml files and mzML files
• Issues with Score
• Run Name is missing
• <Seq> is missing

this one shouldn't be in DB because the sequences are missing
(related #78 (comment))

[colin-combe]

re. PXD021417 - maybe lets leave this in for testing purposes