Skip to main content
. 2022 Jan 6;10(1):116. doi: 10.3390/biomedicines10010116

Figure 4.

Figure 4

Prediction models built based on clinical features and genome-wide SNPs to classify clopidogrel-resistant patients using machine learning methods. (A) Machine learning workflow to select reliable and predictable features to build prediction models for group classification. (B) Accuracy and AUC of SVM model with different feature selection numbers. (C) Pie charts indicating the genotype frequencies of SNPs identified by AI-assisted analysis using SNP datasets obtained from patients of the four groups. # indicates the signaling of SNP array was lower than the calling rate. ** p < 0.005 by Chi-square test. # indicates that the signaling by the SNP array was lower than the calling rate. GS1-279B7.1 is annotated as a pseudogene. SLC37A2, Solute Carrier Family 37 Member 2; IQSEC1, IQ Motif Additionally, Sec7 Domain ArfGEF 1; PSD3, Pleckstrin Additionally, Sec7 Domain Containing 3; BTBD7, BTB Domain Containing 7; GLIS3, GLIS Family Zinc Finger 3.