Section: Application Domains

Gene expression

Participant : Marc Lavielle.

Mixed effects models can also be successfully used in quantitative biology for modeling the dynamics of biological networks in cell populations. Indeed, the population approach is relevant for building predictive computational models of intracellular processes. Popix was interested with the experiments performed by the CONTRAINTES Inria team looking at the high-osmolarity glycerol (HOG) pathway in budding yeast. Yeast cells are exposed to osmotic shocks, i.e., sudden changes in the solute concentration of their surroundings. Signal transduction pathways, most notably the HOG pathway, provide information to the cell about the osmolarity of its environment and activate responses to deal with these stress conditions. In particular, a large set of genes is turned on and corresponding stress-responsive proteins are produced. This protein production process can be quantified by replacing one target protein, for example STL1, by a fluorescent protein such as yECitrine. This can be done by genetically modifying the yeast genome.

Thanks to time-lapse microscopy and cell tracking algorithms, single cell responses can be measured over time. Significant inter-cell variability is often observed.

The related Hog1-induced gene expression model is given by a parametric reaction network. Monolix can then be used to estimate the model parameters.

A collaboration with LIFEWARE (formerly CONTRAINTES) is starting on this subject.