Exploiting the determinants of stochastic gene expression in Saccharomyces cerevisiae for genome-wide prediction of expression noise

摘要: Gene regulation is a process with many steps allowing for stochastic biochemical reactions, which leads to expression noise—i.e., the cell-to-cell fluctuation in protein abundance. Such noise can give rise drastically diverse phenotypes, even within isogenic cell populations. Although numerous biophysical approaches had been proposed model origin and propagation of biological networks, these models essentially characterize innate dynamics gene mechanistic way. In this work, by investigating context yeast cellular we place formulism onto solid genetic ground. At sequence level, show that extremely noisy genes are highly conserved their coding sequences. level networks where natural selection manifested topological constraints, varying modularly organized interaction network positioned orderly regulatory network. We demonstrate constraints predictive expression, were able confidently predict more than 2,000 whose was previously not known. validated predictions high-content imaging. Our approach makes feasible genome-wide prediction such predictability turn suggests an evolvable trait.