Variant detection and variant analysis process for diagnosis of CH and MODY

DESCRIPTION (provided by applicant): Correlagen develops and commercializes DNA-based diagnostic testing for monogenic diseases, such as congenital hyperinsulinism (CH) and maturity-onset diabetes of the young (MODY), in the areas of endocrinology and immunology. Currently, sequence variant detection is based on PCR amplification and full sequencing of gene coding regions and intron/exon junctions. The first specific aim is to expand on this variant detection methodology to allow detection of genomic deletions in addition to sequence variation. Recent studies have shown that genomic deletions cause disease phenotypes at a significant frequency {Eichler, 2006; McCarroll, 2006; Walsh, 2006). Such deletions are not detectable by direct Sanger DNA sequence analysis of genomic DNA if they span an entire PCR-amplification fragment (amplicon) since sequence is still obtained from the other chromosome copy for autosomal genes. Thus, variant detection by DNA sequencing alone can potentially lead to a false negative result. This limitation is of particular concern for the diagnosis of diseases commonly associated with large deletions in a gene, such as MODY5 (Bellanne-Chantelot, 2005). Therefore, a standardized quantitative PCR protocol for deletion detection will be developed that can be easily integrated with Correlagen's existing full-sequencing platform. The second specific aim focuses on refining Correlagen's current variant scoring method to create a new, multi-factorial scoring protocol. This advanced scoring protocol will be used to analyze variants discovered in CH and MODY genes during diagnostic testing. A number of these variants are newly discovered and are currently classified as variants of unknown significance, limiting their diagnostic value. Various existing algorithmic-based methods for interpreting variant significance will be analyzed for their capacity to predict the known function of variants accurately and for patterns leading to false predictions. A prevalence study on DNA samples derived from the general population will be performed to identify common polymorphisms in CH and MODY genes, and calculator for assessing the strength of genotype/phenotype correlations will developed. Based on these studies, new algorithms and genetic parameters for evaluating variants will be developed. Concurrently, an information technology platform will be developed to track changes in scores as the scoring protocol evolves. The improved variant detection and variant scoring protocols resulting from the studies proposed in specific aims 1 and 2 will be incorporated into Correlagen's current service processes to enhance the clinical utility of its testing services for the diagnosis of genetic diseases. Our efforts will lead to a systematic and reliable method of variant detection and analysis that can be applied to variants found in other genes tested by Correlagen and by other reference laboratory licensees of our technology. More broadly, these methods also have the potential to become standard methodology for genetic testing performed in both academia and industry.