Preeclampsia Susceptibility Assessment Based on Deep Learning Modeling and Single Nucleotide Polymorphism Analysis

Area of Science:

• Genetics
• Immunology
• Reproductive Medicine

Background:

• Early diagnosis of preeclampsia remains a significant challenge in improving pregnancy outcomes.
• Interleukin-13 (IL-13) and Interleukin-4 (IL-4) pathways are implicated in inflammatory processes relevant to pregnancy complications.

Purpose of the Study:

• To investigate the potential of the IL-13 and IL-4 pathway in early preeclampsia detection.
• To analyze the association between IL-13 gene polymorphisms (rs2069740 and rs34255686) and preeclampsia risk.
• To develop a combined diagnostic model for early preeclampsia detection.

Main Methods:

• Utilized microarray data (GSE149440) to construct an expression matrix.
• Applied Gene Set Enrichment Analysis (GSEA) to identify IL-13 and IL-4 pathway genes.
• Developed deep learning models (multilayer perceptron, PPI graph convolutional neural network) based on gene expression levels.
• Genotyped IL-13 polymorphisms (rs2069740, rs34255686) using Amplification Refractory Mutation System (ARMS) PCR.

Main Results:

• IL-4 and IL-13 pathway gene expression levels significantly differentiated early preeclampsia from normal pregnancies.
• Significant differences in genotype distribution and allelic frequencies for IL-13 rs34255686 and rs2069740 were observed between preeclampsia cases and controls.
• Identified specific risk markers associated with these polymorphisms.

Conclusions:

• The expression levels of IL-13 and IL-4 pathway genes hold significant potential for early preeclampsia diagnosis.
• IL-13 gene polymorphisms rs2069740 and rs34255686 are associated with preeclampsia risk.
• A combined diagnostic model integrating genetic polymorphisms and gene expression data could be developed for future clinical application in preeclampsia diagnostics.