Extended Data Fig. 1: A uniformly processed resource of CF–MS data.

a, Approaches to protein quantification employed by published CF–MS experiments. SILAC, stable isotope labelling by amino acids in cell culture; iBAQ, intensity-based absolute quantification. b, Proportion of the organismal proteome quantified in each CF–MS experiment (grey lines, individual datasets; blue line, mean across all datasets). c, Cumulative distribution of the number of proteins quantified per dataset in between one and 25 fractions. d, GO term enrichment among CORUM proteins detected in at least one CF–MS fraction, left, or never detected, right. e, PaxDb consensus protein abundance of mouse proteins detected or never detected by CF–MS. f, Coverage of high, moderate, and low abundance proteins (expressed as a mean proportion of fractions in which these proteins were detected) in published human CF–MS experiments (n = 46). g, As in f, but with CF–MS experiments divided into three groups based on the length of the liquid chromatography gradient. ***, p < 0.001, two-sided Spearman rank correlation. h, PaxDb consensus protein abundance of human proteins in the CORUM database and non-CORUM proteins. i, Difference in the number of protein groups quantified in each CF–MS experiment, compared to the processed chromatogram data accompanying the original publications (grey lines, individual datasets; blue line, mean across all datasets).