Date Published: November 20, 2009
Publisher: Public Library of Science
Author(s): Osamu Ogasawara, Kousaku Okubo, Justin C. Fay. http://doi.org/10.1371/journal.pone.0007943
Abstract: The relative contributions of natural selection and random genetic drift are a major source of debate in the study of gene expression evolution, which is hypothesized to serve as a bridge from molecular to phenotypic evolution. It has been suggested that the conflict between views is caused by the lack of a definite model of the neutral hypothesis, which can describe the long-run behavior of evolutionary change in mRNA abundance. Therefore previous studies have used inadequate analogies with the neutral prediction of other phenomena, such as amino acid or nucleotide sequence evolution, as the null hypothesis of their statistical inference.
Partial Text: It has long been hypothesized that phenotypic evolution is more often based on evolutionary changes in gene expression regulation than on sequence changes in proteins . Prompted by this hypothesis and the advent of genome-wide gene expression profiling techniques, an increasing number of studies have investigated the pattern of evolutionary change in gene expression profiles and the evolutionary forces governing the process.
In this study, we introduce a novel, explicitly defined model of gene expression evolution. Our model assumes that (1) the mRNA abundance is affected by mutations proportional to the abundance before mutation; (2) the number of expressed genes and total number of mRNA molecules in a cell type is nearly constant throughout a specific evolutionary process; (3) the mRNA abundance of each gene has a lower limit near 1.0 copy/cell; and (4) the mRNA abundance of each gene (fi) suitable for its function is determined by its relative relationship to other genes, i.e., fi>fj.