EVENT : C3BI Seminars
Main speaker : Julien Dutheil, from Institute of Evolutionary Sciences, University of Montpellier & Max Planck Institute for Evolutionary Biology, Plön Date : 15/06/2017 at 02:00 pm Location : Retrovirus room – LWOFF (22) ,Institut Pasteur, Paris
Between chaos and determinism: the evolution of stochastic gene expression
While often depicted as finely tuned programs, biochemical processes in the cell are subject to random fluctuations because of the small number of molecules involved. While this phenomenon was long recognised, it was only recently assessed at a genomic scale. In particular, the advent of single cell transcriptomics technologies have provided us with a picture of the amount of each mRNA present in a cell giving us access to a cell-specific measure of gene expression. The comparison of multiple, isogenic cells allows us to measure the variance in gene expression (also called “noise”). Just as average gene expression, expression noise is also subject to natural selection. Housekeeping genes, for instance, have been reported to be less noisy. Conversely, genes involved in stress or immune response are more variable in their expression, an observation which was hypothesised to result from a bet-edging strategy. In order to quantify transcriptional noise, we used single-cell transcriptomes of mouse cells. I will discuss statistical biases in the current measures of noise and introduce a new statistic for its estimation. We compared gene-specific transcriptional noise with other genomic variables, including sequence conservation and 3D genome architecture. Our results demonstrate that the major factors explaining transcriptional noise are linked to gene network architecture, consistent with a multi-level selection model where selection acts at the pathway level to prevent noise propagation.
Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting