Table 4 Cancer-control classification performance on held-out data with area under the precision-recall curve (AUPRC) after feature selection by different methods
From: Finding associations in a heterogeneous setting: statistical test for aberration enrichment
Features | Top 300 Limma | Top 300 heterogeneous | Top Het. features | All features | Literature |
|---|---|---|---|---|---|
| Â | features | features | transformed | Â | features |
Ovarian cancer | 0.695 | 0.743 | 0.948 | 0.696 | 0.503 |
Breast cancer | 0.530 | 0.612 | 0.965 | 0.632 | 0.541 |