DNA-methyleringsmarkers voor zwangerschapshypertensie via machine learning

BACKGROUND:This study aims to develop a prediction model to identify individuals at risk of hypertensive disorders of pregnancy (HDPs), including gestational hypertension and preeclampsia, by integrating epigenetic biomarkers and clinical factors in the first trimester of pregnancy.METHODS:A 2-stage nested case-control study, matched by age and body mass index, was conducted with 618 pregnant women in China, with peripheral blood samples collected in the first trimester to evaluate the average methylation levels of differentially methylated regions (DMRs) between controls and HDP cases. In stage 1 (discovery set), 24 controls and 27 cases were used to identify the differential DMRs. In stage 2, 294 controls and 273 cases were used to validate the previously identified DMRs. DMRs selected from the intersectional results of lasso regression, XGBoost, random forest, and Shapley Additive Explanations models were further combined with women’s clinical risk factors to construct prediction mo