Stochasticity or Noise in Biochemical Reactions

Heterogeneity in gene expression across isogenic cell populations can give rise to phenotypic diversity, even when cells are in homogenous environments. This diversity arises from the discrete, stochastic nature of biochemical reactions, which naturally arise due to the very small numbers of genes, RNA, or protein molecules in single cells. Modern measurements of single biomolecules have created a vast wealth of information about the fluctuations of these molecules, but a quantitative understanding of these complex, stochastic systems requires precise computational tools. In this article, we present modern tools necessary to model variability in biological system and to compare model results to experimental data. We review the Chemical Master Equation and approaches to solve for probability distributions of discrete numbers of biomolecules. We discuss how to fit probability distributions to single-cell data using likelihood based approaches. Finally, we provide examples of fitting discrete stochastic models to singlemolecule fluorescent in-situ hybridization data that quantifies RNA levels across populations of cells in bacteria and yeast.