Classification and Regression Tree Model for the Differential Diagnosis of Preeclampsia Based on Clinicopathological Features and miR Signatures

Preeclampsia (PE) is a pregnancy complication characterized by the onset of high blood pressure after 20 weeks of gestation with proteinuria and abnormal liver enzymes. The early diagnosis and prophylactic use of aspirin can reduce the long-term complications of PE. In the current study, we utilized machine learning tools for the differential diagnosis of EOPE and LOPE based on demographic, clinical, and biochemical data. We employed SYBR green-based real-time PCR to study the differential expression of hsa-miR-4743-5p, miR-149-5p, miR-331-5p, and miR-483-5p in both forms of PE.A classification and regression tree (CART) model was developed to differentiate between EOPE and LOPE.This was achieved by determining thresholds of systolic blood pressure (SBP), Diastolic Blood Pressure (DBP), Mean Arterial Pressure (MAP), Body Mass Index (BMI), urine protein, and SGOT.The RT-PCR-based DEM profile identified an association of miR-4743-5p with both forms of PE; with LOPE.MiRDip analysis revealed that genes targeted by these miRs influence TGF beta signaling in EOPE; cholesterol and lipid homeostasis and NOTCH2 signaling in LOPE.In conclusion, SBP, MAP, BMI, urine protein, DBP, and SGOT are key determinants of EOPE and LOPE.The DEM profile clearly distinguished EOPE and LOPE.