BRCA1-specific machine learning model predicts variant pathogenicity with high accuracy.

Identification of novel BRCA1 variants outpaces their clinical annotation which highlights the importance of developing accurate computational methods for risk assessment. Therefore our aim was to develop a BRCA1-specific machine learning model to predict the pathogenicity of all types of BRCA1 variants and to apply this model and our previous BRCA2-specific model to assess BRCA variants of uncertain significance (VUS) among Qatari patients with breast cancer. We developed an XGBoost model that utilizes variant information such as position frequency and consequence as well as prediction scores from numerous in silico tools. We trained and tested the model with BRCA1 variants that were reviewed and classified by the Evidence-Based Network for the Interpretation of Germline Mutant Alleles (ENIGMA) consortium. In addition we tested the model's performance on an independent set of missense variants of uncertain significance with experimentally determined functional scores. The model performed excellently in predicting the pathogenicity of ENIGMA-classified variants (accuracy: 99.9%) and in predicting the functional consequence of the independent set of missense variants (accuracy: 93.4%). Moreover it predicted 2 115 potentially pathogenic variants among the 31 058 unreviewed BRCA1 variants in the BRCA exchange database. Using two BRCA-specific models we did not identify any pathogenic BRCA1 variants among those found in patients in Qatar but predicted four potentially pathogenic BRCA2 variants, which could be prioritized for functional validation.

Prevalence and Genotype-Phenotype Correlation of Lynch Syndrome in a Selected High-Risk Cohort from Qatar's Population.

Lynch syndrome (LS) is the most common cause of hereditary colorectal cancers (CRC) and is associated with an increased risk for ovarian and endometrial cancers. There is lack of knowledge on the epidemiology of LS in the non-Caucasian populations especially in Qatar. The aim of this retrospective study is to explore the prevalence of LS in a selected high-risk cohort in the State of Qatar in addition to investigating the frequency and genotype-phenotype correlation associated with mismatch repair genes pathogenic variants. Retrospective review of medical records of 31 individuals with LS, 20 affected with colorectal cancer and 11 unaffected with family history of cancers, referred from January 2017 until August 2020. The prevalence of LS among affected and unaffected patients is 22% (20/92) and 2.2% respectively. Among affected individuals, MLH1 and MSH2 genes were highly frequent while for unaffected individuals, a recurrent PMS2 pathogenic variant was reported in several related individuals suggesting a tribal effect. This study highlights the epidemiology of LS in high-risk cohort in Qatar which helps to provide recommendations on genetic testing, and personalize surveillance and management programs.