![]() 20–23 HCM is common enough to provide the large datasets needed for these gene-specific and data-driven approaches. It has been suggested that disease and gene-specific approaches are needed to improve interpretation, 19 and guidelines have been produced for specific genes and/or disease areas. ![]() In our cohort, eight sarcomeric genes collectively provide firm molecular diagnoses for ~27% of HCM patients, with a further ~13% of patients carrying a VUS in the same genes. Hypertrophic cardiomyopathy (HCM), a relatively common autosomal dominant disease (1 in 500 prevalence), is a major cause of heart disease in people of all ages 18 and a cause of sudden cardiac death. However, wherever large patient cohorts are attainable, mutational hotspots and the uncertainty surrounding in silico predictors can be directly estimated from the data. Furthermore, although much work has gone into the development of in silico prediction scores, alternative scores can be conflicting, leading to discordance among testing laboratories 17 and uncertainty in their application (criteria PP3: ‘Multiple lines of computational evidence support a deleterious effect on the gene or gene product’). Although positional information is covered by criteria PM1 (‘Located in a mutational hot spot and/or critical and well-established functional domain (eg, active site of an enzyme) without benign variation’), there is a lack of robust statistical evidence for mutational hotspots, resulting in inconsistent application of this criterion. However, due to limited information available for many variants, they fall into the category ‘variant of uncertain significance’ (VUS). 16 These guidelines integrate diverse data and classify variants into five categories from benign to pathogenic. The American College of Medical Genetics and Genomics (ACMG) has produced guidelines to interpret variant pathogenicity. Unlike previous approaches to address this problem, 7 15 we present computationally fast methods, for a realistic Mendelian disease genetic model, that place equal weight on the burden and clustering signals, making it a viable alternative strategy where simple burden testing has been unsuccessful. Here, we detect association based on a dominant model of rare deleterious variants and demonstrate that power can be increased by including variant residue position alongside gene-level burden. Furthermore, there are no compelling examples where rare variants play a protective role. ![]() 13 14 However, due to sample size limitations, few methods exist to test the rare disease ultra-rare variant hypothesis in a case–control setting. Several enhancements to this simple approach have been developed including weighting by variant frequency or functional annotation, 11 integrating additional genetic risk factors such as polygenic risk scores 12 or modelling both protective and deleterious variants by comparing variance in variant-level case–control frequencies. The aggregated burden of rare variants in affected cases compared with healthy controls has proved to be a useful test to confirm candidate 9 and identify novel, 10 putative pathogenic genes. 8 With technical advances in high-throughput, exome sequencing has become another approach to identify novel pathogenic genes and variants. Mendelian disease genes were historically identified by linkage and candidate gene studies in multiplex affected families. 6 Despite numerous examples of variant clustering, there have been few attempts to explicitly model variant residue position as a predictor of pathogenicity. ![]() 1–5 A plausible mechanism underpinning this phenomenon is the presence of multiple loss or gain-of-function variants within functionally important domains. The clustering of pathogenic missense variants in specific regions or domains of proteins has been frequently reported.
0 Comments
Leave a Reply. |