Research Article
Multiclass Prediction with Partial Least Square Regression for Gene Expression Data: Applications in Breast Cancer Intrinsic Taxonomy
Table 1
Performance of PLS-regression classifiers for prototypical arrays.
| Intrinsic subtype | Basal-like | HER2-enriched | Luminal-A | Luminal-B | Normal breast-like |
| Number of samples | 57 | 35 | 23 | 12 | 12 | PLS-regression | | | | | | Number of gene component | 1 | 1 | 2 | 1 | 2 | -variance explained | 57.0% | 37.1% | 74.5% | 25.8% | 60.2% | -variance explained | 86.7% | 56.2% | 64.6% | 24.6% | 66.5% | Binary LR | | | | | | Adjusted -square | 0.99 | 0.73 | 0.9 | 0.63 | 0.99 | AUC | 1 | 0.96 | 0.99 | 0.96 | 1 | Accuracy | 98.6% | 89.9% | 97.1% | 95.0% | 100.0% | Sensitivity | 98.2% | 74.3% | 91.3% | 50.0% | 100.0% | Specificity | 98.8% | 95.2% | 98.3% | 99.2% | 100.0% |
|
|
PLS: partial least square, LR: logistic regression, AUC: area under the curve.
|