Establishment of an Artificial Neural Network Model Using Immune-Infiltration Related Factors for Endometrial Receptivity Assessment

A comprehensive clinical strategy for infertility involves treatment and, more importantly, post-treatment evaluation. As a component of assessment, endometrial receptivity does not have a validated tool. This study was anchored on immune factors, which are critical factors affecting embryonic implantation. We aimed at establishing novel approaches for assessing endometrial receptivity to guide clinical practice.

M1/M2 polarization influences endometrial receptivity by regulating three gene modules, while the established ANN model can be used to effectively assess endometrial receptivity to inform pregnancy and individualized clinical management strategies.

Immune-infiltration levels in the GSE58144 dataset (n = 115) from GEO were analysed by digital deconvolution and validated by immunofluorescence (n = 23). Then, modules that were most associated with M1/M2 macrophages and their hub genes were selected by weighted gene co-expression network as well as univariate analyses and validated using the GSE5099 macrophage dataset and qPCR analysis (n = 19). Finally, the artificial neural network model was established from hub genes and its predictive efficacy validated using the GSE165004 dataset (n = 72).

Dysregulation of M1 to M2 macrophage ratio is an important factor contributing to defective endometrial receptivity. M1/M2 related gene modules were enriched in three biological processes in macrophages: antigen presentation, interleukin-1-mediated signalling pathway, and phagosome acidification. Their hub genes were significantly altered in patients and associated with ribosomal, lysosomal, and proteasomal pathways. The established model exhibited an excellent predictive value in both datasets, with an accuracy of 98.3% and an AUC of 0.975 (95% CI 0.945-1). Conclusions: M1/M2 polarization influences endometrial receptivity by regulating three gene modules, while the established ANN model can be used to effectively assess endometrial receptivity to inform pregnancy and individualized clinical management strategies.