Figure 6.
Screening of core genes and validation based on machine learning methods in RA and MDD. (A) Coefficient profiles of variables in the LASSO regression model in RA. (B) Ten-fold cross-validation for turning parameter (λ) selection in the LASSO regression model in RA. (C) The optimum root mean squared error (RMSE) of SVM-based method based on 44 characteristic genes in RA. (D) Coefficient profiles of variables in the LASSO regression model in MDD. (E) Ten-fold cross-validation for turning parameter (λ) selection in the LASSO regression model in MDD. (F) The optimum root mean squared error (RMSE) of SVM-based method based on 27 characteristic genes in MDD. (G) 14 core genes in RA and 14 core genes in MDD screened by LASSO regression model and SVM-based method, and 6 common core genes were obtained after taking the intersection.
