Dynamic chromosome conformation signatures as prognostic and diagnostic disease biomarkers and their relationship to the non-coding nascent transcriptome

We are interested in the stable fractal-like folded network adopted by eukaryotic genomes.  We use two approaches to uncover these folded states. First is a proprietary machine learning algorithm that assigns potential contact points in the genome, which are regions with a high propensity to fold into structures but share no obvious primary sequence similarity or feature. An experimental step, based on an optimised 3C (capturing chromosome conformation) technique, detects >2x106 long-range interactions (16kb to 300kb) between subsets of potential contact points in blood chromatin from individuals. Interestingly, we have discovered that these structures switch from one stable state to another in diseases, such as cancers, immune conditions, and metabolic disease, and this finding has been developed into robust, novel, diagnostic and prognostic tests, based on the EpiSwitchTM platform, by the Oxford University spin-out Oxford BioDynamics plc. Very little is understood about the underlying mechanisms that facilitate formation of one particular stable fractal state, or the switching of genome structures from one state to another. In this seminar, I will discuss our new method, known as size-selected native elongating transcript sequencing (ssNET-Seq), for defining novel lowly expressed non-coding transcription units throughout the yeast genome and show that they correlate with the sites of loop anchors and our new thinking on the relationship between genotype and phenotype.