PPM2-0207-170010

Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in women in their reproductive years associated with diabetes, gestational diabetes, infertility and endometrial cancer. We estimate using data from women in the Qatar Biobank that the prevalence of PCOS is 21%, and we have shown that of those patients with type 2 diabetes in the age range 18-40 years that there is double the incidence of diabetes compared to a Caucasian population. Currently there is difficulty in making the diagnosis of PCOS in adolescence and there are no current markers either for the definitive diagnosis of PCOS or to predict those who will develop associated comorbid conditions. PCOS demonstrates non-Mendelian familial aggregation that is consistent with a complex genetic disease resulting from the interaction between susceptibility genes and the environment[1]. It has been suggested from twin studies that the heritability is around 79% from twin studies. Two genome wide association studies in Chinese populations have identified 11 risk loci for PCOS that count for 17 single nucleotide polymorphisms (SNPs) that are associated with the disease. A number of these were found to be common to a European GWAS study but others were not. These SNPs have been related to glucose and lipid metabolism as well as ovarian hormonal regulation and cell cycle regulation. Only a small amount of the heritability of PCOS is accounted for by the common susceptibility variants mapped so far, around 5%. However, whilst these markers have been confirmed as associated with the condition, there have not been any detailed studies using deep genome sequencing in pedigree studies that should enrich the sample for causal common and especially rare variants. The combination of high heritability and high prevalence has great potential for novel findings using large Qatari pedigrees showing excess aggregation of PCOS. Hypothesis: Rare or less common variants (<5%) may responsible for a significant proportion of PCOS. These variants have somewhat larger effect sizes and can best be detected in pedigree studies where the frequencies of the rare alleles are enriched, especially if there are multiple variants occurring within the same gene as is common with many diseases. Barriers to detecting these rare variants include technical issues, but with the use of next-generation sequencing technologies and novel analytical strategies this can now be overcome. We propose an approach that examines shared genomic segments (SGS) obtained from whole-genome capture and NextGen resequencing. This method identifies SGS that harbor rare or even private pedigree-specific variants, in addition to common variants. We will apply this initial statistical approach to 15 large, high-risk extended pedigrees with 15+ meioses among distantly related subjects. Using the shared segments identified, we will use a new powerful statistical approach using the program pVAAST, which combines association, linkage, conservation scores and polymorphism functional assessment into one overall likelihood statistic. This method identifies the most likely functional rare variants in SGS-identified shared regions that may not be captured by common variants from the usual GWAS approach. The pedigree-based studies increase the power to identify variants with low frequencies (<5%) because these pedigrees were selected to segregate PCOS, increasing the likelihood that the same variant is responsible in most, if not all, affected members within a pedigree. Requiring that candidate causal variants cosegregate in pedigrees is an effective way to exclude false positive findings and is one of the strengths of our study. We will then extend and validate the significant pedigree findings in the Qatar Biobank PCOS subjects with whole genome sequencing available. Using the larger Qatari Biobank sample will provide the opportunity to evaluate allele frequencies of the identified variants, and to validate the variants that are not private to a specific pedigree in a more general population. Common variants will also be evaluated for replication of prior studies or for identification of new loci more specific for the Qatari population. Family members from PCOS clinics will form a prospective validation on newly presenting patients with PCOS diagnosed according to the Rotterdam criteria (4 in total) so that they can be stratified into the differing phenotypes to see if the association of genetic variants with this expanded definition of PCOS is even stronger. Specific aims. 1. Identify common and especially rare variants/private (pedigree specific) mutations (<0.5%) that contribute to PCOS. Select 15 extended pedigrees segregating PCOS that contain as many as possible female relatives with PCOS with at least 15 meioses between them. Power to detect true rare variants exceeds 80% for this sample. Apply whole genome capture and NextGen resequencing to the 120 PCOS individuals. Identify shared genomic segments for each pedigree and the underlying shared potential causal variants. 2. Validate the sequences derived from the sequencing approach described in 1 using the whole genome sequencing that is available for the Qatar Biobank. This project will be undertaken by Dr Stephen Atkin who is an international expert in PCOS, Dr Steven Hunt who is an international expert in statistical genetics and Dr Ibrahim Janahi who is a senior Qatari consultant in endocrinology with a very large Qatari practice based in Hamad General Hospital. The successful outcome of this study will be to identify both rare and common genetic variants that make up the hereditability of PCOS in Qatar and then validate them using the Qatar Biobank. This will then facilitate defining diagnostic criteria for addressing the heterogeneity that is seen both for the expression of the differing phenotypes but also of the associated conditions related to PCOS.