mzmlstats allows the user to produce a file containing all features for every signal in the MS/MS experiment. The produced file is a parquet file, with the original name of the file plus the following postfix {file_name}_ms_info.parquet. Here, the definition of each column and how they are estimated and used:

• scan: The scan accession for each MS and MS/MS signal in the mzML, depending on the manufacturer, the scan will have different formats. Example, for thermo (e.g controllerType=0 controllerNumber=1 scan=43920). We tried to find the definition of quantms.io.
• ms_level: The MS level of the signal, 1 for MS and 2 for MS/MS.
• num_peaks: The number of peaks in the MS. Compute with pyopenms with spectrum.get_peaks().
• base_peak_intensity: The max intensity in the spectrum (MS or MS/MS).
• summed_peak_intensities: The sum of all intensities in the spectrum (MS or MS/MS).
• rt: The retention time of the spectrum, capture with pyopenms with spectrum.getRT().

For MS/MS signals, we have the following additional columns:

• precursor_charge: The charge of the precursor ion, if the signal is MS/MS. Capture with pyopenms with spectrum.getPrecursors()[0].getCharge().
• precursor_mz: The m/z of the precursor ion, if the signal is MS/MS. Capture with pyopenms with spectrum.getPrecursors()[0].getMZ().
• precursor_intensity: The intensity of the precursor ion, if the signal is MS/MS. Capture with pyopenms with spectrum.getPrecursors()[0].getIntensity(). If the precursor is not annotated (present), we use the purity object to get the information; see note below.
• precursor_rt: The retention time of the precursor ion, if the signal is MS/MS. See note below.
• precursor_total_intensity: The total intensity of the precursor ion, if the signal is MS/MS. See note below.

NOTE: For all the precursor-related information, we are using the first precursor in the spectrum. The following columns intensity (if not annotated), precursor_rt, and precursor_total_intensity we use the following pyopnems code:

precursor_spectrum = mzml_exp.getSpectrum(precursor_spectrum_index)
precursor_rt = precursor_spectrum.getRT()
purity = oms.PrecursorPurity().computePrecursorPurity(precursor_spectrum, precursor, 100, True)
precursor_intensity = purity.target_intensity
total_intensity = purity.total_intensity

mzmlstats allows the user to produce a file containing all the MS2 spectra including the intesities and masses of every peak. The produced file is a parquet file, with the original name of the file plus the following postfix {file_name}_ms2_info.parquet. Here, the definition of each column and how they are estimated and used:

• scan: The scan accession for each MS and MS/MS signal in the mzML, depending on the manufacturer, the scan will have different formats. Example, for thermo (e.g controllerType=0 controllerNumber=1 scan=43920). We tried to find the definition of quantms.io.
• ms_level: The MS level of the signal, all of them will be 2.
• mz_array: The m/z array of the peaks in the MS/MS signal. Capture with pyopenms with mz_array, intensity_array = spectrum.get_peaks().
• intensity_array: The intensity array of the peaks in the MS/MS signal. Capture with pyopenms with mz_array, intensity_array = spectrum.get_peaks().

We use the FeatureFinderMultiplexAlgorithm from OpenMS to extract the features from the MS1 spectra. We use an algorithm based on the original implementation by Andy Lin. The output of this algorithm is a feature map, which contains the following information:

• feature_mz: The m/z of the feature.
• feature_rt: The retention time of the feature.
• feature_intensity: The intensity of the feature.
• feature_charge: The charge of the feature.
• feature_quality: The quality of the feature.
• feature_percentile_tic: The percentile of the feature in the total ion current.
• feature_id: The unique identifier of the feature generated by OpenMS.
• feature_min_rt: The minimum retention time of the feature within the feature map.
• feature_min_mz: The minimum m/z of the feature within the feature map.
• feature_max_rt: The maximum retention time of the feature within the feature map.
• feature_max_mz: The maximum m/z of the feature within the feature map.
• feature_num_scans: The number of scans that the feature is present in the feature map.
• feature_scans: The scans where the feature is present in the feature map.

The tool will generate a gzip compressed parquet file with the extension {file_name}_ms1_feature_info.parquet.