The logic of gene control by DNA regulatory elements

We are interested in how genomic and epigenetic information is integrated with extrinsic signals to promote concerted changes in gene expression.

Cells in the body share the same genotype, but express an amazing diversity of phenotypes in time and space. Global analyses of transcription factor binding, chromatin state and long-range chromatin interactions have streamlined the annotation of DNA regulatory modules – ‘switches’ of gene expression. However, we still know rather little about how exactly these modules work. In particular, it remains to be understood how the input from intrinsic transcription factors and extrinsic signals, is integrated on the DNA sequence, and how multiple regulatory elements jointly regulate their target genes.

The ‘TF collective’ model of enhancer organisation (Junion/Spivakov et al., 2012) in comparison with other proposed models – ‘billboard’ (Kulkarni and Arnosti, 2003) and ‘enhanseosome’ (Thanos and Maniatis, 1995).


Chromosomal interactions (shown as arcs) of SOX2 gene promoter in human embryonic stem cells (upper) and neural progenitor cells (lower) detected using Promoter Capture Hi-C. Also shown is information on gene expression (mRNA-seq), the patterns of histone modifications, and chromatin states defined jointly on their basis (active chromatin, green; poised chromatin, orange; Polycomb-repressed, red; intermediate, yellow; background, grey). Adapted from Freire-Pritchett et al., eLife 2017.

We aim to decipher the ground rules of gene regulation and establish their functional interplay in biological phenomena involving global changes in phenotype, such as in cell differentiation and activation. Our current focus is on the role of non-coding DNA elements such as enhancers in integrating and transmitting gene regulatory information.

We combine experimental and computational approaches to study these questions, capitalising on our previous work on promoter-enhancer relationships, organisation of DNA regulatory elements and population genomics. Our particular interest is in human primary cells as models, and in genetic and epigenetic variation as “natural perturbations” in the system.

Our ultimate goal is to generate comprehensive functional models of gene control “logic” underlying cellular decisions. Interrogation and validation of these models will pinpoint key individual players (regulatory elements, genes, extrinsic signals) and their regulatory relationships in these processes and shed light on how they are remodelled in disease.