The number of transcriptomics studies has rocketed in recent years, but the weak correlation between transcript and protein abundance means that transcript profiles are of limited use for understanding variation in protein levels. Existing proteomics methods are not powerful enough to monitor proteome changes across many samples: the concentration of a protein can be compared across experiments but, in practice, corresponding proteins in different experiments are difficult to match. The new, more direct approach also relies on the established technique of mass spectrometry (which separates peptides according to mass and charge), but introduces a key innovation: an algorithm that aligns peptides across samples — this made it possible to compare the levels of 569 proteins in more than 400 samples.
The samples in question derive from the parents and offspring of a cross between two strains of Saccharomyces cerevisiae. Protein levels in the segregants varied continuously, that is, they behaved like quantitative traits. To determine the QTLs that control protein abundance, 221 of the most reliable peptides were chosen for linkage studies. Perhaps surprisingly, most control loci seem to function in trans, as they map to chromosomes other than those that contain the genes encoding the corresponding proteins. Furthermore, the regulatory QTLs are clustered in four 'hotspots', one of which affected the abundance of a staggering 35 proteins.
This is a preview of subscription content, access via your institution
