DIA USI example for PXD019909

[ypriverol]

Here a file with all the USIs corresponding to the reanalysis performed using quantms and DIANN workflow.

PXD019909.1-USI.txt.zip

[edeutsch]

As discussed, all these USIs have an errant "sequence=" in them. That needs to be removed to make them valid.

My other concern is that all of the ones that I spot checked look a lot like random hits, so I am concerned that maybe the scan numbers are wrong or something is amiss with the generation of these?

[edeutsch]

One suggestion was to look at all the ids from one scan and see if they use different peaks, etc. I picked on at random:

mzspec:PXD019909:20180914_QE8_nLC0_BDA_SA_DIA_Keratinocytes_NN002:scan:15718:VTSLPDNHK/2
mzspec:PXD019909:20180914_QE8_nLC0_BDA_SA_DIA_Keratinocytes_NN002:scan:15718:NSGSVNMGSR/2
mzspec:PXD019909:20180914_QE8_nLC0_BDA_SA_DIA_Keratinocytes_NN002:scan:15718:GFQVAPEHHNDHK/3
mzspec:PXD019909:20180914_QE8_nLC0_BDA_SA_DIA_Keratinocytes_NN002:scan:15718:AVTIHDTEK/2

These are all spectra for one scan in your list.
Viewing each of these yields results that don't seem at all convincing.

For example for this one:
https://proteomecentral.proteomexchange.org/usi/?usi=mzspec:PXD019909:20180914_QE8_nLC0_BDA_SA_DIA_Keratinocytes_NN002:scan:15718:NSGSVNMGSR/2
All the peaks are in the grass except for y1 (which is shared everywhere and not diagnostic) and b5++. strong b5++ would be diagnostic if it were true. But looking at what a Q Exactive spectrum of this peptide looks like:
https://proteomecentral.proteomexchange.org/usi/?usi=mzspec:PXD003028:20150414_QEp1_LC7_NiGr_SA_Saliva_1_fractionated2:scan:5392:NSGSVNMGSR/2
(from PeptideAtlas)

there is no b++. y5 and y8 should be diagnostic monster peaks. They are nowhere to be found. I think
mzspec:PXD019909:20180914_QE8_nLC0_BDA_SA_DIA_Keratinocytes_NN002:scan:15718:NSGSVNMGSR/2
is a false positive or a malformed USI (I hope the latter)

[edeutsch]

as another point, this USI looks pretty much as good as the original one:
https://proteomecentral.proteomexchange.org/usi/?usi=mzspec:PXD019909:20180914_QE8_nLC0_BDA_SA_DIA_Keratinocytes_NN002:scan:15738:NSGSVNMGSR/2

All I did was arbitrarily add 20 to the scan number. The USI looks equally poor as the original.

[ypriverol]

@edeutsch here is an Excel with all the USIs from the experiment PXD019909 with only one run:

• Project run: https://ftp.pride.ebi.ac.uk/pub/databases/pride/resources/proteomes/quantms-benchmark/PXD019909-singlerun-results/

Excel attached with Spectronaut and quantms USIs:
SPvsDIANN (2).xlsx