Weighted gene co-expression network analysis identifies a novel immune-related gene signature and nomogram to predict the survival and immune infiltration status of breast cancer

Breast cancer is one of the most common cancers in the world and seriously threatens the health of women worldwide. Prognostic models based on immune-related genes help to improve the prognosis prediction and clinical treatment of breast cancer patients. In the study, we used weighted gene co-expression network analysis to construct a co-expression network to screen out highly prognostic immune-related genes. Subsequently, the prognostic immune-related gene signature was successfully constructed from highly immune-related genes through COX regression and LASSO COX analysis. Survival analysis and time receiver operating characteristic curves indicate that the prognostic signature has strong predictive performance. And we developed a nomogram by combing the risk score with multiple clinical characteristics. CIBERSORT and TIMER algorithms confirmed that there are significant differences in tumor-infiltrating immune cells in different risk groups. In addition, gene set enrichment analysis shows 6 pathways that differ between high- and low-risk group. The immune-related gene signature effectively predicts the survival and immune infiltration of breast cancer patients and is expected to provide more effective immunotherapy targets for the prognosis prediction of breast cancer.