Abstract
Purpose:
The objective of this research is to pinpoint potential diagnostic markers for endometriosis and explore the immune infiltration patterns linked with this condition through the utilization of machine learning techniques.
Methods:
A total of five gene expression datasets (GSE7305, GSE7307, GSE25628, GSE23339, and GSE120103) were examined in order to identify differentially expressed genes (DEGs) that distinguish normal patients from those with endometriosis. The algorithms Random Forest and Lasso regression were utilised to identify diagnostic biomarkers. GSEA and Go&KEGG database were utilised to determine the potential pathway in which the biomarker was implicated. With the ailment. Furthermore, an assessment of immune cell infiltration in endometriosis tissues relative to normal tissues was conducted using CIBERSORT analysis. In order to investigate the relationship between diagnostic markers and immune cell populations, a correlation analysis was performed.
Results:
DLK1 (Delta-like 1 homolog) has emerged as a potential diagnostic biomarker for endometriosis, with indications suggesting that Notch signalling could be pivotal in the development of endometriosis.
Conclusion:
DLK1 emerges as a promising diagnostic biomarker for endometriosis, as our study indicates a complex interplay between immune dysregulation and disease pathogenesis. Notably, our findings elucidate that DLK1 regulates endometriosis through Notch signaling, highlighting the potential of Notch signalling as a therapeutic target for future interventions.