Skip to main content

Table 1 GA selected PCs and the classification model performance (ROC)

From: Selection of microbial biomarkers with genetic algorithm and principal component analysis

Data for creating PCA

Result

Model

_6 PCs

Model

_5 PCs

Model

_4 PCs

Model

_3 PCs

Model

_2 PCs

Model

_1 PC

All

PCs selected

PC1+, PC2–, PC7+, PC11+, PC15–, PC27–

PC1+, PC2–, PC7+, PC11+, PC27–

PC1+, PC2–, PC7+, PC27–

PC1+, PC2–, PC7+

PC1+, PC7+ (or PC2–)

PC1+

AUC (CV)

0.87

0.85

0.84

0.81

0.77

0.69

Obesity

PCs selected

PC2–, PC4–, PC14+, PC16–

PC18+, PC19–

PC2–, PC4–, PC14+, PC18+

PC19–

PC2–, PC4–, PC14+, PC18+

PC2–, PC14+, PC18+

PC14+, PC18+

PC14+

AUC (CV)

0.92

0.92

0.90

0.87

0.84

0.80

Healthy

PCs selected

PC1+,PC3+, PC5–,PC23+, PC28–,PC34+

PC1+,PC3+, PC23+,PC28–

PC34+

PC1+,PC23+, PC28–,PC34+

PC1+,PC23+, PC34+

PC1+,PC34+

PC1+

AUC (CV)

0.92

0.90

0.88

0.87

0.83

0.72

  1. + Positive correlation coefficient in the model
  2. – Negative correlation coefficient in the model