Table 3:
Assessment of prevalence for each preprocessing method based on the compendium of 30 experiments
|
t2-statistic |
Bayesian network |
|||||
|---|---|---|---|---|---|---|
| FDR | 0.01 | 0.05 | 0.25 | 0.01 | 0.05 | 0.25 |
| PLIER—no MM | 4.03 | 3.83 | 3.48 | 3.95 | 3.23 | 2.72 |
| MBEI—no MM | 4.08 | 3.78 | 3.52 | 4.23 | 3.53 | 3.35 |
| fRMA—default | 4.53 | 4.63 | 4.43 | 4.00 | 3.93 | 3.70 |
| RMA—default | 4.92 | 4.60 | 4.73 | 4.92 | 5.05 | 4.55 |
| RMA—no BG | 5.22 | 5.00 | 4.65 | 4.82 | 5.13 | 4.42 |
| MBEI—default | 4.63 | 4.77 | 4.48 | 5.03 | 5.02 | 5.62 |
| PLIER—default | 5.37 | 5.23 | 4.93 | 6.12 | 6.67 | 6.82 |
| GCRMA—default | 5.83 | 6.50 | 7.50 | 5.57 | 5.62 | 6.87 |
| MAS5.0—default | 6.38 | 6.65 | 7.27 | 6.37 | 6.82 | 6.97 |
| CDF of highest ranked method | ||||||
| P(rank ≤ 1) | 0.46 | 0.52 | 0.48 | 0.48 | 0.67 | 0.93 |
| P(rank ≤ 2) | 0.83 | 0.89 | 0.93 | 0.78 | 0.97 | 1.00 |
| P(rank ≤ 3) | 0.94 | 0.96 | 0.98 | 0.99 | 1.00 | 1.00 |
| P(rank ≤ 4) | 0.99 | 0.99 | 1.00 | 1.00 | 1.00 | 1.00 |
| P(rank ≤ 5) | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
The upper portion of the table shows the average rank of the prevalence scores across the compendium of datasets. A lower rank corresponds to a larger number of gene sets called differentially expressed. The lower portion of the table displays the CDF of the highest ranked method, providing an assessment of our confidence in the highest ranking. Note: MM, mismatch; BG, background correction