Mammalian genomes contain tens of millions of cytosine bases that have been modified to form methylcytosine, which in some instances can regulate local transcriptional activity. It has been observed that when proteins, such as transcription factors, bind to DNA, the local chromatin state is altered and methylcytosines within the vicinity can become demethylated.
Consequently, the millions of methylcytosines throughout the genome may be considered as individual sensors for local protein binding activity. Comphrehensive maps of DNA methylation state throughout the genome, which can now be routinely generated with existing high throughput DNA sequencing technologies, thus constitute an information rich datatype from which protein binding activity could potentially be detected from DNA alone. This would provide a means to reconstruct complex genome regulatory processes when only genomic DNA is available. This project aims to develop effective computational approaches for reconstructing the chromatin structure of the genome solely from DNA methylation maps, in order to explore the complex regulatory processes that govern gene expression patterns in diverse human cell types, development, and in health and disease states.
Professor Ryan Lister – [email protected]