Speaker
Description
Cells have several possible fates (diseases, differentiation, death). A way to describe those is via ‘cell trajectories’, in which the transcriptome of a cell evolves from a state A to a state B in a defined manner. Understanding transcriptome evolution is critical for identifying or improving treatments for various diseases. Here, we chose to study a particular cellular fate: differentiation. To understand this mechanism, we aim to predict the real-time evolution of the transcriptome in reprogrammed stem cells as they differentiate into neuronal cells via neural stem cells. We develop a new mathematical approach to identify genes that regulate neuronal differentiation from RNA sequencing data, using a workflow recently developed by our team. It involves three steps. First, the selection of bimodal genes, i.e., those with two expression modes (high or low) within the same population, as differentiation is a switch phenomenon. Then, the cells were clustered based on bimodal gene expression patterns. Finally, applying differential expression analysis to clustered data, we can identify genes that are over- or underexpressed between clusters. These correspond to genes that regulate neuronal differentiation. Starting from a sample consisting of two replicates of 19500 genes and 13000 cells each, we can isolate the 500 most relevant genes for cell differentiation. It will lead to the construction of gene regulatory networks and the development and analysis of deterministic models.
