Interplay of digital and analog control in time-resolved gene expression profiles

Abstract

Background: Measuring the agreement between a gene expression profile and a known transcriptional regulatory network is an important step in the functional interpretation of bacterial physiological state. In this way, general design principles can be explored. One such interpretive framework is the relationship of digital control, that is, the impact of sequence-specific interactions, and analog control, i.e., the extent of the influence of chromosomal structure. Methods and Results: Here, we present time-resolved gene expression profiles of Escherichia coli's growth cycle as measured by RNA-seq. We extend methods which have been developed for discrete sets of differentially expressed genes and apply them to the wild type and two mutant time-series for which the global transcriptional regulators fis and hns were inactivated. We test our continuous methods using simulated 'expression profiles' generated from random Boolean network dynamics where we observe a clear trade-off between maximum response and level of detail included. In the real time-course expression data, we find strong interdependent changes of digital and analog control during the exponential growth phase and a dominance of analog control during the stationary phase. Conclusions: Our investigation puts forward a simple and reliable method for quantifying the match between time-resolved gene expression profiles and a transcriptional regulatory network. The method reveals a systematic compensatory interplay of digital and analog control in the genetic regulation of E. coli's growth cycle.