The study investigates the relationship between protein interaction networks and evolutionary rates in yeast. It shows that proteins with more interactions evolve more slowly, not because they are more important to the organism, but because a greater proportion of their structure is involved in function. This is supported by data showing a significant negative correlation between the number of interactors and evolutionary rate. The study also explores two hypotheses for this correlation: one involving structural constraints limiting substitutions, and another involving stronger purifying selection due to higher fitness effects. However, the data suggest that the correlation is not mediated by fitness effects, but rather by coevolution, where substitutions in one protein lead to reciprocal changes in interacting partners. The study confirms that interacting proteins evolve at similar rates, as predicted by coevolution. Further analysis shows that the similarity in evolutionary rates is not due to similar fitness effects, but rather to coevolution. The study also highlights the potential application of these findings in assessing the quality of large interaction data sets through genomic sequence comparisons. The research provides insights into the genetic architecture of natural variation in gene expression, showing that over 1500 genes are differentially expressed between strains, with 570 genes linked to loci affecting their expression. The study identifies eight trans-acting loci, each affecting the expression of multiple genes. Overall, the findings contribute to understanding the evolutionary dynamics of protein interaction networks and the genetic basis of natural variation in gene expression.The study investigates the relationship between protein interaction networks and evolutionary rates in yeast. It shows that proteins with more interactions evolve more slowly, not because they are more important to the organism, but because a greater proportion of their structure is involved in function. This is supported by data showing a significant negative correlation between the number of interactors and evolutionary rate. The study also explores two hypotheses for this correlation: one involving structural constraints limiting substitutions, and another involving stronger purifying selection due to higher fitness effects. However, the data suggest that the correlation is not mediated by fitness effects, but rather by coevolution, where substitutions in one protein lead to reciprocal changes in interacting partners. The study confirms that interacting proteins evolve at similar rates, as predicted by coevolution. Further analysis shows that the similarity in evolutionary rates is not due to similar fitness effects, but rather to coevolution. The study also highlights the potential application of these findings in assessing the quality of large interaction data sets through genomic sequence comparisons. The research provides insights into the genetic architecture of natural variation in gene expression, showing that over 1500 genes are differentially expressed between strains, with 570 genes linked to loci affecting their expression. The study identifies eight trans-acting loci, each affecting the expression of multiple genes. Overall, the findings contribute to understanding the evolutionary dynamics of protein interaction networks and the genetic basis of natural variation in gene expression.