Figure 3.

Construction and validation of the Lasso regression signature. (a) The lowest point of the lasso regression curve determines the penalty value. (b) Depending on whether the penalty value intersects each curve, the lncRNAs in the signature were selected and we calculated the regression coefficients. (c) The forest map is constructed based on the univariate Cox regression analysis. (d) The area under the curve (AUC) value of the signature at different prediction endpoints using the receiver operating characteristic (ROC) curve.