Table 2 Results of DNN on TARGET dataset.
From: Network-based integration of multi-omics data for clinical outcome prediction in neuroblastoma
Dataset | Feature type | ACC (%) | F1 score | ROC AUC | Feature dim |
|---|---|---|---|---|---|
RNA-Seq | Centrality | 59.4 ± 6.3 | 0.69 ± 0.05 | 0.62 ± 0.07 | 13 |
Modularity | 61.6 ± 6.8 | 0.67 ± 0.09 | 0.62 ± 0.07 | 64 | |
Both | 60.7 ± 3.5 | 0.64 ± 0.09 | 0.64 ± 0.04 | 77 | |
Abridged | 61.5 ± 4.6 | 0.66 ± 0.1 | 0.66 ± 0.05 | 38 | |
Methylation | Centrality | 57.1 ± 6.5 | 0.36 ± 0.25 | 0.54 ± 0.09 | 13 |
Modularity | 54.1 ± 4.2 | 0.49 ± 0.19 | 0.51 ± 0.03 | 22 | |
Both | 54.6 ± 1.6 | 0.31 ± 0.11 | 0.52 ± 0.06 | 35 | |
Abridged | 54.1 ± 0.1 | 0.51 ± 0.13 | 0.53 ± 0.06 | 34 | |
Network-level fusion | Centrality | 57.2 ± 5.2 | 0.57 ± 0.07 | 0.62 ± 0.08 | 13 |
Modularity | 64.8 ± 8.7 | 0.65 ± 0.17 | 0.69 ± 0.07 | 44 | |
Both | 65.1 ± 4.7 | 0.68 ± 0.04 | 0.71 ± 0.06 | 57 | |
Abridged | 64.0 ± 5.0 | 0.69 ± 0.05 | 0.77 ± 0.07 | 11 | |
Feature-level fusion | Centrality | 55.7 ± 3.0 | 0.45 ± 0.29 | 0.55 ± 0.02 | 13 |
Modularity | 60.5 ± 7.9 | 0.67 ± 0.05 | 0.61 ± 0.10 | 55 | |
Both | 61.1 ± 8.0 | 0.58 ± 0.12 | 0.61 ± 0.07 | 68 | |
Abridged | 56.8 ± 8.1 | 0.64 ± 0.12 | 0.60 ± 0.15 | 39 |
