Ulysse Herbach will present his recent work the 5th of April at 2pm

Gene network inference from single cell data: a mechanistic approach

The behaviour of a cell population largely depends on decisions made at the individual level, resulting from a complex interaction network between genes within each cell. Recent technologies make it possible to get some traces (mRNA levels) of the activity of this network in individual cells, where biological stochasticity is preserved from averaging. Though it is still a challenging task, taking full advantage of this type of data could take gene network inference (the problem of reconstructing the network from the generated traces) to the next level.

In this talk, I will focus on the construction of a stochastic dynamical model for a set of interacting genes, in which each link between two genes is directed and has an explicit interpretation in terms of chemical reactions. An interesting point is that the traditional framework for gene regulation can be seen as a deterministic approximation of this model. In parallel, a mean-field approximation allows to compute explicitly the network stationary distribution: this second approach can be used to derive a promising statistical model for the data, where stochasticity is not just noise but also contains information.