Modeling gene expression: noise and cooperativity

GUTIERREZ, P.S.; CALCAGNI LAURA; MONTEOLIVA DIANA; DIAMBRA, L.

Montevideo

Congreso; ISCB LATIN AMERICA 2010; 2010

International Society for Computational Biology

All chemical reactions have intrinsic fluctuations that are inversely proportional to the systemsize. Fluctuations in gene expression are particularly pronounced because of the small copynumber of species undergoing transitions between discrete chemical states and the small size ofbiological compartments [1]. At transcriptional level, gene expression is mainly controlled bytranscription factors (TFs) that bind specifically at upstream promoter sites and influence thebinding of other molecules to cis-regulatory system (CRS).We propose a stochastic model for gene expression regulation that include several binding sites,as show in Figure 1. By means of master equation formalism and computer simulation, our modelpredicts that the Hill coefficient and the level of noise increases as the interaction energy betweenactivators increases [2].The model allows to distinguish between two cooperative binding mechanisms, which we studyseparately: one of them increases the binding probability of regulatory proteins, the otherincreases the stability of the TF-DNA bound. While expression levels are not affected differentlyby what mechanism is acting, they induce very different fluctuation levels in the expression.When the transition between the states of the regulatory system is quick in relation to theproduction and degradation rates of transcript, we found that the first mechanism is associatedto an unimodal distribution of the transcript number, while in the second case it yields a bimodaldistribution.