Endometriosis-related alterations in the endometrium revealed by integrated single-cell and AI-powered approaches.

Endometriosis, affecting 1 in 9 women, presents treatment and diagnostic challenges. To address these issues, we generated a comprehensive single-cell atlas of endometrial tissue, comprising 466,371 cells from 35 endometriosis and 25 non-endometriosis donors without exogenous hormonal treatment. Detailed analysis reveals significant gene expression changes and altered receptor-ligand interactions present in the endometrium of endometriosis patients, including increased inflammation, adhesion, proliferation, cell survival, and angiogenesis in various cell types. These alterations may enhance endometriosis lesion formation and identify potential therapeutic targets. Using ScaiVision, we trained neural network models to predict endometriosis of varying disease severity (median AUC = 0.83), including one model based solely on a set of 11 genes confirmed as dysregulated in endometriosis patients through differential expression analysis. In conclusion, our findings reveal numerous pathway and ligand-receptor changes in the endometrium of endometriosis patients, offering insights into pathophysiology, potential targets for improved treatments, and predictive models for enhanced outcomes in endometriosis management. Our models, while not yet externally validated, can serve as a tool for hypothesis generation and starting point for further clinical development.