Enhanced interpretation

We will undertake a comprehensive analysis of the clinical features through comparison of HPO terms to determine if unrelated individuals could have the same diagnosis. This will allow comparison of whole genome data i.e. to determine if variants in the same gene ares hared in unrelated individuals or by analysis of trios (unaffected parents and children) to identify spontaneous (de novo) causative variants.

We will identify families with rare diseases with the following criteria for analysis. No known diagnosis following initial 100k Genomes analysis; Affected siblings (multiplex family) OR consanguineous parents OR a single affected individual with a well-defined phenotype suggestive of a specific autosomal-recessive disorder. We expect that this will include ~1500 – 1800 cases recruited across England. We will then prioritise cases for study based on recognition of shared clinical features. In cases where there is a known clinical diagnosis and only one or no pathogenic variants in the known causative gene has been identified we will screen the data for intronic variants and undertake RNA based analysis or determine if structural variants are present. For consanguineous families we will prioritise the analysis of novel or ultra rare homozygous variants – initially coding variants but then non-coding variants in or near attractive candidate genes based on known function.

It is clear that genome sequence data is of high quality in detecting single nucleotide variants (SNVs) and that validation is not technically challenging. However, working with colleagues across all GMCs we will consider complex mutational mechanisms e.g. structural variants like inversions, intronic variants that create cryptic splice sites to determine criteria around the optimum approaches for validation. These will result in best practice guidelines to ensure optimal clinical testing strategies.

Previously we and others have demonstrated that systematic re-analysis of the genomic data can help in the identification of novel disorders and reveal new disease mechanisms. In this project we will analyse genomic data of patients recruited into the rare diseases arm of the 100,000 Genomes project. The patients for this study will be selected on the basis of their genotype e.g. variants in a specific gene or pathway. The genes will be selected on the basis of their involvement in a particular pathway, process or knowledge of other related genes associated with inherited disorders. We expect that the phenotypes will be very broad and span a range of phenotypes across different age ranges and so it is not possible or appropriate to focus on a single clinical participant set. With recruiting physicians, we will undertake reverse phenotypic (assess whether certain features perhaps originally not reported are present - reverse phenotyping). Where relevant, for selected variants/genes of interest we will interrogate the100,000 Genomes Data to identify additional patients from other GMCs.