Fig. 5

Machine learning for screening biomarkers. A: LASSO Logic Coefficient Penalty Graph; Each curve represents the variation trajectory of each independent variable coefficient, the y-axis represents the coefficient value, and the upper x-axis represents the number of non-zero coefficients in the model; B: LASSO Logic Coefficient Penalty Graph; The horizontal axis represents log (Lambda), while the vertical axis represents the error of cross validation; C: Support Vector Machine Model Accuracy (Left) and Error Rate (Right); D: Top 10-variable importance. “mean Decrease accuracy “indicates the degree of reduction in the accuracy of random forest prediction. The higher the value represents the greater the importance of the variable; “mean decrease Gini “calculates the impact of each variable on the heterogeneity of observations at each node of the classification tree. The higher the value represents the greater importance of variables; E: Lollipop map of top10 characteristic gene; F: Venn diagram for selecting biomarkers