We study the mechanisms by which programs of gene expression are selected and perpetuated during the development of multicellular organisms. Regulatory sequence elements contain clusters of binding sites for transcription factors, most of which interact with some specific DNA sequence motif. By discovering the repertoire of transcription factor binding sites, we can uncover an important part of the cell’s regulatory network. We are addressing this question using a new computational motif discovery tool, NestedMICA, to find DNA sequence motifs that are over-represented in larger sets of regulatory sequences from across the genomes of a panel of multicellular organisms.

It has become increasingly clear that the function of regulatory elements depends on their context in terms of nuclear location and chromatin structre. To this end, we are keen to understand the landscape and functions of stable epigenetic modifications - particularly DNA cytosine methylation. High-throughput sequencing technologies allow epigenetic marks to be studied on a genome-wide basis, and we have used a combination of deep sequencing and a new analytical technique to generate the first map of DNA methylation across a complete vertebrate genome. We are now combining this technology with other analysis and data visualisation methods in order to study how DNA methylation interacts with other regulatory and epigenetic mechanisms. We are also investigating how human DNA methylation changes are associated with ageing and complex diseases.

A regulatory motif discovered in the Drosophila genome, and the embryonic expression pattern of a gene regulated by this motif. (P Tomancak et al. Genome Biology 3:research0088)

The BioTIFFIN interface for browsing regulatory sequence motifs.

The Methyl DNA Immunoprecipitation (MeDIP) technique can be used to quantify the methylation state of genomic DNA on a large scale. In methylated regions (coloured blue), signal correlates with the density of CpG dinucleotides..

Plain English:
Genome sequences are available for many organisms, but tools for finding and interpreting the sequences which regulate gene activity remain very limited. My group is developing new computational techniques for analysing gene regulators. Gene regulators consist principally of clusters of short DNA words (motifs), each of which enables the regulation of a gene by one of several hundred transcription factors. Therefore, we are creating comprehensive catalogs or dictionaries of regulatory motifs found in the human genome, and in several key model organisms. We then use them to annotate complete genome sequences and understand which genes are controlled by which transcription factors.

Understanding gene regulation will help to answer many fundamental biological questions, especially in developmental biology where little is known about how the many cell types found in complex organisms differentiate from stem cells. It has been shown that some genetic diseases, including one form of the blood disease thalassemia, are caused by defects in regulators rather than protein-coding genes. Accurate and comprehensive identification of gene regulators is essential to fully understand genetic disease and variation between individuals.

Selected publications:

• Rakyan VK, Down TA, Maslau S, Andrew T,Yang T-P, Beyan H,Whittaker P, McCann OT, Finer S,Valdes AM, Leslie RD, Deloukas P, and Spector TD (2010) Human aging- associated DNA hypermethylation occurs preferentially at bivalent chromatin domains Genome Research 20:434-439

• Kolasinska-Zwierz P, Down T, Latorre I, Liu T, Liu XS and Ahringer J (2009) Differential chromatin marking of introns and expressed exons by H3K36me3 Nature Genetics 41:376-381

• Down TA, Rakyan VK,Turner DJ, Flicek P, Li J, Kulesha E, Graf S, Johnson N, Herrero J,Tomazou EM,Thorne NP, Backdahl L, Herberth M, Howe KL, Jackson DK, Miretti MM, Marioni JC, Birney E, Hubbard TJP, Durbin R,Tavare S and Beck S (2008) A Bayesian deconvolution strategy for immunoprecipitation-based DNA methylome analysis Nature Biotech 26:779-785

• Down TA, Bergman CM, Su J and Hubbard TJP (2007) Large scale discovery of promoter motifs in Drosophila melanogaster PLoS Comput Biol 3:e7

• Down TA and Hubbard TJP (2005) NestedMICA: sensitive inference of over-represented motifs in nucleic acid sequences. Nucleic Acids Res 33, 1445-1453

Resources:

• BioTIFFIN regulatory motif database
• NestedMICA motif-discovery toolkit

Methyl-DNA immunopreciptation (MeDIP-seq) data before and after normalisation using the Batman method. Viewed in the Dalliance genome visualisation tool

Multiple epigenetic marks “paint” exons in the genome. In the case of DNA methylation (shown here) the marking of exon boundaries is remarkably sharp, and appears to be independent of transcription. (single-base methylation data from Lister et al, Nature 462, 315-322).