Exploration of driver biomarkers and immune microenvironment in patients with endometriosis: Evidence from RNA-seq and machine learning

Endometriosis (EM) is a condition that impacts roughly 10% of women within the reproductive age demographic on a global scale. Due to the limitations of conventional diagnostic techniques for endometriosis, a concerted effort is necessary to improve the existing diagnostic and develop novel diagnostic tools or biomarkers. With improvements in gene sequencing methods and a steady drop in sequencing prices, more endometriosis-associated genes have been discovered recently. Utilizing a random forest classifier, we conducted a screening of differentially expressed genes (DEGs) pertaining to EM sourced from the Gene Expression Omnibus (GEO) database, and six key genes were screened out (DDX56, TAL1, ALX3, DDX6, ADRBK2, and ZMYND11). The DDX6 and ADRBK2 protein expression changes in the eutopic endometrium of the different groups were evaluated by immunohistochemical methods to further validate their diagnostic value. To verify this method, we selected 20 clinical samples for testing, and the test results were exactly the same as the clinical report. The contribution of all these genes associated with endometriosis has never been investigated. Additionally, we have developed a novel diagnostic model for endometriosis that uses an artificial neural network and have conducted successful tests on publicly available datasets to evaluate its diagnostic performance. Numerous studies have unveiled important roles for immunity in the pathogenesis of EM; thus, we also assessed the content of immune cells in the eutopic endometrium from different samples. Compared to normal samples, the proportion of CD8 + T cells, T follicular helper cells, and monocytes in the eutopic endometrium of women with EM were significantly higher, while the proportion of gamma-delta T cells, macrophages, dendritic cells resting, and mast cells resting was significantly lower. The immune factor is a major determinant of the course of the disease, may be valuable prognostic markers and are worth further examination.