Mass Spectrometry Data Import
Loading mzML/mzXML Files with pyOpenMS
from pyopenms import MSExperiment, MzMLFile, MzXMLFile
exp = MSExperiment()
MzMLFile().load('sample.mzML', exp)
for spectrum in exp:
if spectrum.getMSLevel() == 1:
mz, intensity = spectrum.get_peaks()
elif spectrum.getMSLevel() == 2:
precursor = spectrum.getPrecursors()[0]
precursor_mz = precursor.getMZ()
Loading MaxQuant Output
import pandas as pd
protein_groups = pd.read_csv('proteinGroups.txt', sep='\t', low_memory=False)
# Filter contaminants and reverse hits
protein_groups = protein_groups[
(protein_groups['Potential contaminant'] != '+') &
(protein_groups['Reverse'] != '+') &
(protein_groups['Only identified by site'] != '+')
]
# Extract intensity columns
intensity_cols = [c for c in protein_groups.columns if c.startswith('LFQ intensity')]
intensities = protein_groups[['Protein IDs', 'Gene names'] + intensity_cols]
Loading Spectronaut/DIA-NN Output
diann_report = pd.read_csv('report.tsv', sep='\t')
# Pivot to protein-level matrix
protein_matrix = diann_report.pivot_table(
index='Protein.Group', columns='Run', values='PG.MaxLFQ', aggfunc='first'
)
R: Loading with MSnbase
library(MSnbase)
raw_data <- readMSData('sample.mzML', mode = 'onDisk')
spectra <- spectra(raw_data)
header_info <- fData(raw_data)
Missing Value Assessment
def assess_missing_values(df, intensity_cols):
missing_per_protein = df[intensity_cols].isna().sum(axis=1)
missing_per_sample = df[intensity_cols].isna().sum(axis=0)
total_missing = df[intensity_cols].isna().sum().sum()
total_values = df[intensity_cols].size
missing_pct = 100 * total_missing / total_values
return {'per_protein': missing_per_protein, 'per_sample': missing_per_sample, 'total_pct': missing_pct}
Related Skills
- quantification - Process imported data for quantification
- peptide-identification - Identify peptides from raw spectra
- expression-matrix/counts-ingest - Similar data loading patterns
majiayu000