Abstract
BACKGROUND: Premature ovarian failure (POF) is a highly heterogeneous disorder that occurs in 1% of women of reproductive age. Very few causative genes and variants contributing to POF have been detected, and the disease remains incompletely understood. In this study, we used whole exome sequencing (WES) to identify potential causal variants leading to POF. METHODS: WES was conducted to identify variants in 34 Korean patients with POF, alongside 10 normal controls. Detected variants were filtered using a range of characterized bioinformatics analyses, and the machine learning tools, CADD and VEST, were used to predict pathogenic variants that could cause disease. VarSome was used for a comprehensive interpretation of the variants. Potential causal variants finally screened by these analyses were confirmed using Sanger sequencing. RESULTS: We identified nine potential causative variants in genes previously associated with POF in 8 of 34 (24%) Korean patients by WES variant analysis. These potentially pathogenic variants included mutations in the MCM8, MCM9, and HFM1 genes, which are involved in homologous recombination, DNA repair, and meiosis, and are established as causing POF. Using a combination of CADD and VEST, 72 coding variants were also identified in 72 genes, including ADAMTSL1 and FER1L6, which have plausible functional links to POF. CONCLUSIONS: WES is a useful tool to detect genetic variants that cause POF. Accumulation and systematic management of data from a number of WES studies in specialized groups of patients with POF (family data, severe case populations) are needed to better comprehend the genetic landscape underlying POF.