Development and verification of the nomogram for dilated cardiomyopathy gene diagnosis

Abstract

Abstract Dilated cardiomyopathy (DCM) is a primary myocardial disease of unclear mechanism and poor prevention. The purpose of this study is to explore the potential molecular mechanisms and targets of DCM via bioinformatics methods and try to diagnose and prevent disease progression early. We screened 333 genes differentially expressed between DCM and normal heart samples from GSE141910, and further used Weighted correlation network analysis to identify 197 DCM-related genes. By identifying the key modules in the protein–protein interaction network and Least Absolute Shrinkage and Selection Operator regression analysis, seven hub DCM genes (CX3CR1, AGTR2, ADORA3, CXCL10, CXCL11, CXCL9, SAA1) were identified. Calculating the area under the receiver’s operating curve revealed that these 7 genes have an excellent ability to diagnose and predict DCM. Based on this, we built a logistic regression model and drew a nomogram. The calibration curve showed that the actual incidence is basically the same as the predicted incidence; while the C-index values of the nomogram and the four external validation data sets are 0.95, 0.90, 0.96, and 0.737, respectively, showing excellent diagnostic and predictive ability; while the decision curve indicated the wide applicability of the nomogram is helpful for clinicians to make accurate decisions.