Fig. 3
Screening for potential biomarkers of AD using a machine learning algorithm. (A) PPI network diagram; red and green represent upregulated and downregulated genes, respectively. (B) The minimum absolute contraction and selection operator models (LASSO) were used to select the characteristic genes. (C) The SVM-RFE algorithm selected biomarker feature genes. The red circle at the lowest point in the left image indicates a minimum error rate of 0.213 for 36 genes, whereas the red circle at the highest point in the right image represents a maximum accuracy rate of 0.787 for 36 genes. (D) The random forest tree algorithm was used to evaluate characteristic genes. (E) The top 20 most important genes were identified using the random forest tree algorithm. (F) Venn diagram of the three algorithms used to screen genes. The overlapping parts of the three circles represent 13 common genes obtained from the three machine learning algorithms. AD, Alzheimer’s Disease; SVM-RFE, support vector machine recursive feature elimination.
