Identification of Immune-Related Genes and Small-Molecule Drugs in Interstitial Cystitis/Bladder Pain Syndrome Based on the Integrative Machine Learning Algorithms and Molecular Docking

Figure 3

IC characteristic genes selected based on the integrative machine learning algorithms. (a, b) Identification of the IC characteristic genes using LASSO regression algorithm; the optimal lambda was determined as 0.053 after tenfold cross-validation. (c) Identification of the IC characteristic genes using SVM-RFE algorithm; the classifier showed the minimum error when . (d) Identification of the IC characteristic genes using random forest algorithm; the bar plot shows the mean decrease accuracy of each gene; the top 10 most important genes were selected as characteristic genes. (e) The Venn diagram performs the five most important IC characteristic genes.

(a)

(b)

(c)

(d)

(e)