Research Article
A Genomic-Clinicopathologic Nomogram for the Prediction of Lymph Node Invasion in Prostate Cancer
Figure 3
Construction of LASSO models. (a) Selection of the tuning parameter λ in the LASSO model via 10-fold cross-validation in the imbalanced training set. The optimal λ value of 0.0165, with log (λ) = −4.103, was chosen based on minimum criteria. (b) LASSO coefficient profiles of the 22 genes. (c) Selection of the tuning parameter λ in the LASSO model via 10-fold cross-validation in the SMOTE-balanced training set. The optimal λ value of 0.0066, with log (λ) = −5.021, was chosen based on minimum criteria. (d) LASSO coefficient profiles of the 29 genes. LASSO: least absolute shrinkage and selection operator; SMOTE: synthetic minority oversampling technique.
(a) |
(b) |
(c) |
(d) |