Delay stochastic simulation of single-gene expression reveals a detailed relationship between protein noise and mean abundance

Abstract

To study noise in the number of protein molecules produced in gene expression, we use a delayed reaction model coupling transcription and translation to examine nine biochemical factors. Fourteen numerical experiments were performed, which show clearly the linear scaling behavior between the protein variance and the mean. We found that the most dominant noise source comes from promoter fluctuations; in second place is the death-and-birth process of mRNA. At the translational level, either increasing the protein birth initiation frequency or decreasing the protein decay rate raises the noise level. Results obtained from the classical model in the literature, which is a simplified version of our model, agree qualitatively with ours. However, they lack some important features.