Fig. 5: Overall quantitative performance of benchmarking sample-A and sample-B from 3 LC-MS platforms.

Benchmarking samples A and B were prepared containing known ratios of peptide digestions from human urine, yeast and E. coli, resulting in expected peptide and protein ratios of 1:1 (A/B) for human, 1:2 for yeast and 4:1 for E. coli proteins. The samples A and B were analyzed in three technical replicates on three LC-MS platforms Orbitrap Fusion Lumos, Orbitrap Exploris 480, and timsTOF Pro 2 (Lumos, E480, and TIMS for short), respectively. a–c The number of proteins identified for each individual organism from sample-A and sample-B in Lumos, E480, and TIMS, respectively. d, e Log-transformed ratios (log2(A/B)) of proteins plotted over the log-transformed intensity of sample B in Lumos, E480, and TIMS (proteins). Colored dashed lines represented the expected log2(A/B) values for human (green), yeast (orange), and E. coli (purple) proteins. All box plots indicated the median and the first and third quartiles as the box ends. Whiskers were positioned 1.5-times the interquartile range. g–i The distribution of the coefficients of variation (CV) obtained on the quantified protein intensity across the three technical replicates was plotted for each individual organism. All box plots indicated the median and the first and third quartiles as the box ends. Whiskers were positioned 1.5-times the interquartile range. j Differentially expressed protein (DEP) detection for three organisms from Lumos, E480, and TIMS, respectively. Percentages of significantly changed proteins as DEPs over the total number of quantified proteins in 1:2 and 4:1 condition were used to estimate the sensitivity, while those in 1:1 condition were used to estimate the specificity. k Sensitivity and specificity of the DEP analysis based on receiver operating characteristic (ROC) curves. Source data are provided as a Source Data file.