The problem. Proteomics has long lagged genomics in completeness: a single mass-spec run samples only a fraction of the proteins present, and getting near-comprehensive coverage of a cell’s expressed proteome took impractically long. How much of the proteome can you actually see, and how fast?

The idea. Bekker-Jensen and colleagues optimised the shotgun workflow — fractionation, gradient, and instrument settings — to identify close to the full complement of expressed proteins in human cell lines in a tractable amount of measurement time. The contribution is the demonstration and the resulting deep reference proteome, not a new algorithm.

Why it matters. It sets the ceiling for what expression-level proteomics can currently resolve — useful context for anyone reasoning about how proteomic and transcriptomic coverage compare. RNA-seq routinely quantifies ~all expressed genes; this is the paper showing proteomics closing that gap. It also produces the kind of deep reference dataset that downstream benchmarks (search engines, quantification) are validated on.

Verdict. A strong methods-and-resource paper rather than a conceptual leap — its value is the coverage benchmark and the dataset. Read it to calibrate expectations: when someone reports “the proteome,” this is roughly how deep the best current shotgun runs reach.