Fig. 2

The distribution of gene lipidomics score in the TCGA GC cohort. A Feature selection with LASSO binary logistic regression model. The left part: The longitudinal solid line represents the partial likelihood deviation ±standard error and the longitudinal dotted line indicates that the best parameter is selected according to the minimum value (left) and 1-SE (right). Lambda is the tuning parameter. The right part: y axis represents Coefficients. Each curve in the graph corresponds to the value of each characteristic regression coefficient varying with the log (Lambda) value. B K-M survival curve of the low- and high- lipid score for TCGA GC patients; C-D The distributions of the lipid score and survival status for GC patients