Methods* | DEG | DEVG | DECG | DEG & DEVG | DEG & DECG | DEVG & DECG | DEG & DEVG & DECG |
---|
Mean of accuracy | 0.8333 | 0.5800 | 0.6831 | 0.8571 | 0.8452 | 0.6359 |
0.8631
|
Std of accuracy | 0.0357 | 0.0460 | 0.1481 | 0.0119 | 0.0238 | 0.1003 | 0.0060 |
- Italic value indicates the best performance in method comparison.
- * DEG means that we used only mDEG score (formula 4) to measure modules and applied these quantified modules for sample-clustering; DEVG means that we used only mDEVG score (formula 5); DECG means that we used only mDECG score (formula 6); DEG & DEVG means that we used the combination of DEG and DEVG; DEG & DECG means that we used the combination of DEG and DECG; DEVG & DECG means that we used the combination of DEVG and DECG; DEG & DEVG & DECG means that we used the combination of all, i.e., DEVC score (formula 7). For each combination, the sample–clustering has been rerun 1000 times, and the mean and variance of accuracies are the final performance of such a strategy.