Spectrum raw files from samples were extracted into ms1 and ms2 files using in-house program RawXtractor or RawConverter (http://fields.scripps.edu/downloads.php) (56), and the tandem mass spectra were searched against UniProt human protein database (downloaded on 25 March 2014) (57) and matched to sequences using the ProLuCID/SEQUEST algorithm (ProLuCID version 3.1) (58, 59), with 50-ppm peptide mass tolerance for precursor ions and 600 ppm for fragment ions. The search space included all fully and half-tryptic peptide candidates that fell within the mass tolerance window with no miscleavage constraint, assembled and filtered with DTASelect2 (version 2.1.3) (60, 61) through Integrated Proteomics Pipeline (IP2 version 3, Integrated Proteomics Applications Inc., CA, USA; www.integratedproteomics.com). To accurately estimate peptide probabilities and false discovery rates (FDRs), we used a target/decoy database containing the reversed sequences of all the proteins appended to the target database (62). Each protein identified was required to have a minimum of one peptide of minimum length of six amino acid residues and to be within 10 ppm of the expected m/z. However, this peptide had to be an excellent match with an FDR less than 0.001 and needed to have at least one excellent peptide match. After the peptide/spectrum matches were filtered, we estimated that the protein FDRs were ≤1% for each sample analysis.