Database searching, data filtering, and quantification

The raw files recorded by MS were converted into mzXML format. Precursors for MS/MS fragmentation were checked for incorrect monoisotopic peak assignments²⁷. All MS/MS spectra were matched against a database encompassing sequences of all proteins in the Uniprot Human (Homo sapiens) database and common contaminants such as keratins using the SEQUEST algorithm (version 28)²⁸. Each protein sequence was listed in both forward and reversed orientations to control and estimate the false discovery rate (FDR) of peptide identifications. The following parameters were used for the database search: 10 ppm precursor mass tolerance; 0.1 Da product ion mass tolerance; full trypsin digestion; up to two missed cleavages; variable modifications: oxidation of methionine (+15.9949); fixed modifications: carbamidomethylation of cysteine (+57.0214), N-terminus and lysine TMT modification (+229.1629).

The target-decoy method was employed to evaluate and further control FDRs of peptide identification^{29, 30}, and linear discriminant analysis (LDA) was utilized to distinguish correct and incorrect peptide identifications based on multiple parameters such as XCorr, ΔCn, and precursor mass error^{27, 31–33}. After scoring, peptides less than six amino acid residues were deleted and peptide spectral matches were filtered to a less than 1% FDR based on the number of decoy sequences in the final data set, then the data set was further filtered to <1% FDR at the protein level.

Quantification of confidently identified peptides was based on the TMT reporter ion intensities in MS²⁸. The isotopic information provided by the company (ThermoFisher) was used to calibrate the measured intensities. The median intensity ratio for each unique peptide in each channel was obtained, and eventually the protein ratio is the median value of all unique peptides for the corresponding protein.