Fig. 4

Screening candidate hub genes for PCOS diagnosis. (a) Based on LASSO logistic regression algorithm to screen diagnostic markers. (b) Application of SVM_RFE for biomarker screening. (c) Venn diagram showed the intersection of diagnostic markers obtained by the two algorithms. (d) Expression level of hub genes (CNTN2, CASR, CACNB3, MFAP2). (e)The ROC curve of the diagnostic efficacy in training set based on SVM_RFE. (f) The ROC curve of the diagnostic efficacy in validation set based on SVM_RFE. (g) The ROC curve of the diagnostic efficacy in training set based on XGBoost. (h) The ROC curve of the diagnostic efficacy in validation set based on XGBoost. SVM-RFE support vector machine recursive feature elimination, LASSO least Absolute Shrinkage and Selection operator, XGBoost extreme gradient boosting, ROC receiver operating characteristic curve