Identification of Potential Biomarkers for Progression and Prognosis of Bladder Cancer by Comprehensive Bioinformatics Analysis

Figure 3

Screening feature genes by machine learning. (a) The Venn diagram of key module genes in the TCGA-BLCA and GSE133624 datasets. (b) Penalty graph of twenty-four characteristic variable coefficients. As the penalty coefficient lambda changes, the coefficients of most variables are compressed to zero. (c) In the LASSO logistic regression model, the best lambda value is selected when the 10-fold cross-validation error is minimal. (d, e) Feature genes were selected with the SVM-RFE algorithm at the optimal point 0.00984. (f) Candidate signature genes expression in the TCGA dataset. (g) Expression of candidate signature genes in the GSE133624 dataset.

(a)

(b)

(c)

(d)

(e)

(f)

(g)