Figure 3. Significant correlation between absolute mRNA and protein abundance. (A) Concordance between transcript abundance and per-protein sequence coverage.

Comparison of peptide coverage per protein (Y-axis) and transcript abundance (X-axis), each of which was calculated as described in Materials and Methods. Each point on the scatterplot corresponds to one of the 1,646 genes whose proteins were detected in the PA. The Spearman correlation coefficient between the two datasets is shown on the graph (R_s = 0.511; P < 10⁻⁷), and the bold grey line represents the correlation squared (R²). (B) Concordance between transcript abundance and per-protein spectral counts. Arithmetic mean of the number of spectra counted per protein (Y-axis; Materials and Methods) was plotted as a function of each protein's cognate transcript abundance (X-axis). (C) Concordance between transcript abundance and proteome coverage. Average mRNA signal intensities for each gene are organized into 100 bins with 100 intensities per bin (i.e. bin 1 = intensity 0−99; bin 2 = 100−199; etc). Note that the first three bins are empty (i.e. transcripts with low intensities were detected neither at the mRNA nor at the protein level). Cumulative proteome coverage (black connected squares; right-hand Y-axis) was calculated by adding the total number of proteins detected per transcript intensity bin as each successively higher bin was added to the analysis. Note that although the analysis was carried out to intensities of 50,000, for brevity we have terminated the graph at 10,000 on the X-axis and 60% cumulative coverage on the right-hand Y-axis , since we observed an increase of new detections of only ∼3% between intensities of 10,000 and 50,000. Total protein count (grey vertical bars; left-hand Y-axis) is represented by the height of each bar, denoting the total number of proteins detected per bin.