MsFraggerToPSM

MsFraggerToPSM turns the psm.tsv result of an MSFragger search, as written by FragPipe, into the .qpsm files that PSMBasedQuantificationTIMs reads. With it, an ion mobility run can be identified in FragPipe and quantified in ProteomIQon, in place of the PeptideSpectrumMatching and PSMStatistics pair. For each row of the tsv the tool rebuilds the modified peptide sequence in ProteomIQon notation, looks it up in the PeptideDB database to get the peptide and modified sequence identifiers, and assigns the row to the MS2 spectrum of the mzlite whose scan time and precursor m/z lie closest to the values in the tsv. The tool has no parameter file.

Inputs and outputs

Flag	Meaning	Comes from
`-i`	one or more `psm.tsv` files, or a directory that is searched for `*.tsv`	MSFragger (FragPipe)
`-ii`	the matching `.mzlite` files, or a directory that is searched for `*.mzlite`	MzMLToMzLiteIonMobility
`-o`	the output directory, created when missing
`-d`	the SQLite peptide database	PeptideDB
`-c`	number of pairs converted at the same time, default 1

All four flags are mandatory. The tsv needs the columns Assigned Modifications, Peptide, Spectrum, Retention, Charge, Calibrated Observed M/Z, Calculated Peptide Mass, Delta Mass, Ion Mobility, Peptide Length, Number of Missed Cleavages, Expectation, Hyperscore, Probability and Protein. Retention is read in seconds and stored in minutes. Files are paired by position, the first tsv with the first mzlite and so on. Give both lists in the same order and with the same length, otherwise the tool converts nothing. The tool stops when the database file does not exist. The lookup uses the unlabeled sequences of the database. Rows whose modified sequence has no entry in the database are dropped, so the database has to be built with the same modifications as the FragPipe search.

From the Assigned Modifications column the tool understands oxidized methionine and carbamidomethylated cysteine, plus N-terminal acetylation. A row with any other modification is printed to the console and dropped.

The tool writes one <run>.qpsm per mzlite into the output directory, named after the mzlite. It is a tab separated table with a header and one row per identified spectrum, with the spectrum identifier of the mzlite, the database identifiers, scan time, charge, precursor m/z, ion mobility, the MSFragger expectation, hyperscore and probability, the sequence and the protein names. An existing .qpsm of the same name is appended to. Delete the old file before a rerun. The output directory also receives MSFraggerToPSM_log.txt.

Running the tool

Install the tool with dotnet tool install --global ProteomIQon.MsFraggerToPSM, then convert one run:

proteomiqon-msfraggertopsm -i path/to/run/psm.tsv -ii path/to/run.mzlite -d path/to/AraTest.db -o path/to/output

Several runs, paired by position and converted three at a time:

proteomiqon-msfraggertopsm -i path/to/run1/psm.tsv path/to/run2/psm.tsv path/to/run3/psm.tsv -ii path/to/run1.mzlite path/to/run2.mzlite path/to/run3.mzlite -d path/to/AraTest.db -o path/to/output -c 3

All flags:

proteomiqon-msfraggertopsm --help