Our genome encodes complex regulatory networks that direct the development of omnipotent cells into many stable differentiated cell types. However, even the regulatory networks underlying differentiated cell types remain surprisingly flexible: many differentiated cell types can be reprogrammed to a pluripotent state, or directly into another differentiated cell type.

The main aim of my group is to develop quantitative models to predict the functional consequences of perturbations of gene regulatory networks, from gene expression to complex phenotypes.  By integrating various genetic and epigenetic sequencing data sets we try to understand how the expression of genes is regulated during differentiation and how genetic variation affects this process.

We are an interdisciplinary group consisting of computational and experimental researchers. This allows us to generate the data we need to build these models, in addition to using existing genomics data sets. We make use of two experimental model systems:

1)   Trans-differentiation of human skin fibroblasts to neurons by overexpression and repression of transcription factors;

2)   Differentiation of human induced pluripotent stem cells to neurons by forced expression of transcription factors.

We use methods from the field of machine learning and statistics, such as Bayesian networks, to construct predictive statistical models. We use high-throughput sequencing techniques such as single-cell RNA-sequencing, transcription factor profiling and epigenetic profiling to measure the state of the regulatory network and the consequences of perturbations.

There are both wet-lab and computational projects for bachelor and master students. I am especially interested to hear from prospective students and postdocs who are interested in combining wet lab molecular biology experiments with quantitative modeling.

Molecular Developmental Biology web page: