Abstract

Recurrent spontaneous abortion (RSA) is a major reproductive challenge with multifaceted causes, including genetic and immunological influences. Complement system-related genes (CRGs) have been implicated in RSA pathogenesis, yet their exact roles remain unclear. This study aims to identify related biomarkers for RSA and investigate the underlying mechanisms. Datasets GSE65099 and GSE26787 were obtained from the GEO database. The GSE65099 dataset was used to identify differentially expressed genes (DEGs), which were intersected with key module genes identified from WGCNA to obtain DE-CRGs related to RSA. Then, key CRGs were screened by combined PPI networks, machine learning, and external validation. A nomogram model was developed for the diagnosis of RSA with key CRG expression levels, followed by validation via ROC curve analysis. Transcription factor-key gene and chemical-gene networks were constructed to reveal the potential mechanisms of key CRGs. Totally, 363 DE-CRGs were selected, and GRAP2 and TREM2 were identified as key genes for RSA, which were both upregulated in RSA samples. The two genes showed excellent diagnostic performance in both the training and validation cohorts. GSEA revealed that GRAP2 and TREM2 were involved in immune-related pathways. Immune infiltration analysis showed that GRAP2 and TREM2 had a negative correlation with effector memory CD4 T cells. Drug prediction identified that the two genes had 3 common interacting chemicals, including bisphenol A, tetrachlorodibenzodioxin, and air pollutants. In conclusion, this study identified GRAP2 and TREM2 as key genes for RSA and provided insights into the immune mechanisms underlying the condition.