Table 2. Comparison of RGBM and RGENIE with inference methods on DREAM5 networks of varying sizes.
| Methods | Data used | DREAM5 experiments | |||||
|---|---|---|---|---|---|---|---|
| Network 1 | Network 3 | Network 4 | |||||
| AU pr | AU roc | AU pr | AU roc | AU pr | AU roc | ||
| RGBM (LS-Boost) | KO,Exp | 0.537 | 0.846* | 0.086 | 0.633* | 0.048 | 0.546 |
| RGBM (LAD-Boost) | KO,Exp | 0.513* | 0.842* | 0.084 | 0.628* | 0.047* | 0.544* |
| ENNET | KO,Exp | 0.432+ | 0.857 | 0.069 | 0.632+ | 0.021 | 0.532+ |
| iRafNet | KO,MTS,Exp | 0.364 | 0.813 | 0.112 | 0.641 | 0.021 | 0.523 |
| RGENIE | Exp | 0.343− | 0.821− | 0.104− | 0.623− | 0.022− | 0.524− |
| GENIE (Winner) | Exp | 0.291 | 0.814 | 0.094 | 0.619 | 0.021 | 0.517 |
| TIGRESS (15) | KO,Exp | 0.301 | 0.782 | 0.069 | 0.595 | 0.020 | 0.517 |
| CLR (18) | Exp | 0.217 | 0.666 | 0.050 | 0.538 | 0.018 | 0.505 |
| ARACNE | Exp | 0.099 | 0.545 | 0.029 | 0.512 | 0.017 | 0.500 |
Here, we provide the mean AUpr and AUroc values for 10 random runs of different inference methods. Here, KO, knockout; KD, knockdown; WT, wildtype; MTS, modified smoothed version of the time-series data; Exp, steady-state gene expression. The best results are highlighted in bold. *, + and − represent the quality metric values where RGBM, ENNET and RGENIE techniques respectively defeat the winner of DREAM5 challenge, i.e. GENIE.