Assigning Significance in Label-Free Quantitative Proteomics to Include Single-Peptide-Hit Proteins with Low Replicates

Figure 5

𝐴_PRO scatter plots, local variability, and thresholds for selecting differentially regulated proteins. The blue dots represent the 𝐴_PRO scatter plot of 𝐴_S versus 𝐴_R corresponding to the unlabeled proteins in sample S_P versus R_P. 𝐴_S is the average of 𝐴_S,1 and 𝐴_S,2. 𝐴_R is the average of 𝐴_R,1 and 𝐴_R,2. The red dots represent the 𝐴_PRO scatter plot of 𝐴_cS versus 𝐴_cR corresponding to the labeled proteins in control sample replicate cS_P versus cR_P. 𝐴_cS is the average of 𝐴_cS,1 and 𝐴_cS,2. 𝐴_cR is the average of 𝐴_cR,1 and 𝐴_cR,2. 𝐴_S,1, 𝐴_S,2, 𝐴_R,1, 𝐴_R,2, 𝐴_cS,1, 𝐴_cS,2, 𝐴_cR,1, and 𝐴_cR,2 were the 𝐴_PRO values for the eight quantitation categories defined in Table 1. To evaluate the local noise of 𝐴_PRO measurement, the relative standard deviation (rSTD) for each protein was calculated from its four unlabeled 𝐴_PRO values 𝐴_S,1, 𝐴_S,2, 𝐴_R,1, and 𝐴_R,2 (the blue trace) or its four labeled 𝐴_PRO values 𝐴_cS,1, 𝐴_cS,2, 𝐴_cR,1, and 𝐴_cR,2 (the pink trace). The rSTD- 𝐴_PRO traces were smoothed with a 100-point moving box. The grey straight lines indicated a 3-fold (solid line) and a 2-fold (dashed line) change threshold. The solid red and green curves represent the fold-change thresholds established with the PLGEM-STN statistics based on the local variance in the null distribution (the pink-dot scatter plot).