One of the most surprising aspects of biology is the diversity of cellular life. Although much of this variability has been attributed to genetic and environmental factors, recent studies have demonstrated4 that genetically identical organisms in the same environment exhibit heterogeneity in gene expression. This phenomenon, gene expression noise, has been observed and measured in species as divergent as prokaryotes and mammalian cells.1-4We have been keenly interested in understanding more about the origin of this heterogeneity and asking what potential functional consequences it may have. We began to address these questions with a series of experiments described in the following chapters. In Chapter 1, we examine the effects that selection can have on gene expression noise in silico. We found that under fairly general conditions, directional selection for the value of a quantitative phenotypic trait can produce increased noise in that trait. In Chapter 2, we will begin to experimentally examine the effects that cis-regulatory mutations have on gene expression noise. . By creating libraries of promoter mutants, we look for mutations that modify the noise; furthermore, by also observing how these same mutations influence the average noise levels of gene expression, we test the independence of these two parameters. By measuring the average and noise levels of gene expression in three mutant yeast promoter libraries, we see that in some cases the two parameters are independent. More importantly, we see a wide variety of mean values ​​and noise, suggesting that cis-regulatory mutations can control gene expression noise independently of the mean. After discovering that specific promoter mutations can impact gene expression noise, we were in..... .the focus of the article ......will help uncover mechanistic insights into the observed fast bursts. To summarize, we examined several potential mechanisms for generating noise in gene expression, namely selection, cis-regulatory mutations, and the localization of trans-acting factors. It seems that everyone can contribute to the noise of gene expression; in particular, the explosion of transcription factors in and out of the nucleus, in addition to leading to an increase in the amount of noise in gene expression, also produces proportional control of a series of genes. These findings also provide a mechanism for proportional control, a phenomenon that may be important in different contexts and may also provide a rational explanation for some of the gene expression disorders we observe. We anticipate that further investigations into the functional consequences of noise will continue to provide insights into cellular behavior.