|
High-throughput protein function, expression, interaction, and localization assays are becoming widespread, producing 1000¢s of systematically measured features of genes, their encoded proteins, and their mutational phenotypes. A central challenge is connecting this growing molecular- and cellular-level information with growing data on genetic variation to interpret their organismal consequences. Using data from quantitative shotgun proteomics, we describe the extent to which protein levels in cells are set by mRNA levels versus other factors, and the large extent to which evolution appears to constrain protein expression levels. These constraints relate to the tendency for proteins to operate not in isolation, but in pathways and complexes, and a systematic analysis of protein complexes reveals an intimate relationship between protein complexes and mutational phenotypes that is likely conserved across evolution. In fact, the organization of proteins into complexes accounts for a large proportion of gene essentiality and can be predictive of organismal phenotypes. Because of these conserved relationships among protein expression, interactions, and mutational phenotypes, phenotypes can be mapped rationally between organisms, revealing new models of disease and candidate disease genes (the phenolog hypothesis), which we demonstrate for the case of new angiogenesis genes. Consideration of the cellular organization of proteins thus leads directly to an explanation of the organismal-level consequences of their perturbation. |
|