*EASTBIO* A quantitative approach to understanding regulation of developmental genes by bivalent promoters

Histone modifications play key roles in the regulation of gene expression. In embryonic stem cells, promoters of developmental genes carry a signature combination of histone H3 lysine 4 trimethylation (H3K4me3), which is usually found at active promoters, and H3K27me3, a hallmark of repressive chromatin. These so-called bivalent domains are thought to poise genes for timely induction during development, but it remains unclear whether and how bivalent domains affect gene expression. One current hypothesis suggests that bivalent domains may act to fine-tune both kinetics and output levels of transcription to ensure accurate temporal and spatial control of expression of key genes during development. However, quantitative kinetic data on bivalent gene expression that would allow to test this hypothesis remains sparse.

In this project, we will utilise an inducible fluorescent reporter gene system to quantify expression parameters (such as latency, activation threshold, maximal output, synchronicity between cells) as a function of epigenetic state. Model bivalent domains with different GC content and CpG density will be integrated upstream of the reporter and their influence on expression will be quantified by Western blot, FACS, and other quantitative readouts. In parallel, epigenetic status of the reporter constructs will be determined using ChIP, ChIP-seq, and related approaches. We will then use mathematical modelling, incorporating available systems-level knowledge on feedback loops connecting Polycomb and Trithorax complexes placing the bivalent modifications and on their interaction with the transcription machinery, to develop a framework that allows modelling gene expression parameters for bivalent domains. We will then use knockouts or CRISPRi of factors involved in the system to test the validity of the model and to generate novel, testable hypotheses regarding functional interactions between the complexes involved and regulatory properties of the bivalent domain module.

This interdisciplinary project will allow training in both experimental chromatin biology as well as bioinformatics and modelling approaches.