CliPPER
CliPPER is an empirical method based on Gaussian graphical models that (i) selects pathways with covariance matrices or means significantly different between two experimental conditions; (ii) on such pathways, identifies the portions of the pathways, called signal paths, mostly associated to the phenotype. The significant pathways reported have both tests significant at alpha level <= 0.05.
| Pathway | alphaMean | alphaVar |
|---|---|---|
| Legionellosis | 0.0 | 0.06 |
| Osteoclast differentiation | 0.03 | 0.28 |
| NF-kappa B signaling pathway | 0.0 | 0.39 |
| Influenza A | 0.02 | 0.38 |
| One carbon pool by folate | 0.5 | 0.0 |
| Type II diabetes mellitus | 0.47 | 0.05 |
| Natural killer cell mediated cytotoxicity | 0.05 | 0.58 |
| Herpes simplex infection | 0.05 | 0.66 |
| Prion diseases | 0.67 | 0.04 |
| Acute myeloid leukemia | 0.71 | 0.04 |
| Salivary secretion | 0.7 | 0.05 |
| Bile secretion | 0.81 | 0.01 |
| Phototransduction | 0.87 | 0.0 |
- alphaVar = adjusted p-value of CliPPER test on the concentration matrices of the pathway between groups.
- alphaMean = adjusted p-values of CliPPER test on the means of the pathways between groups.
