Table 1 Different regression-based methods applied to the time-series gene expression data to construct gene regulatory networks
Method | Data used | Description |
|---|---|---|
iBMA-prior | Gene expression + external data | Our proposed methodology that incorporates prior model probabilities in BMA. These prior probabilities were computed using external data sources. |
iBMA-shortlist | Gene expression + external data | Iterative BMA that uses external knowledge to shortlist p = 100 candidates for each target gene. The revised supervised step was used. Unlike iBMA-prior, the information from the external data is not used in variable selection via BMA. |
Network A from Yeung et al. [3] | Gene expression + external data | This method is the same as in iBMA-shortlist, but using the old version of supervised step described in Yeung et al. [3]. We aim to study the impact of the revised supervised step by comparing iBMA-shortlist to network A. |
LASSO-shortlist | Gene expression + external data | LASSO [36, 63] with the use of external knowledge to shortlist p = 100 candidates for each target gene. |
LAR-shortlist | Gene expression + external data | LAR [64] with the use of external knowledge to shortlist p = 100 candidates for each target gene. |
iBMA-size | Gene expression data only | A simplified version of iBMA-prior that disregards external knowledge, except for setting π gr  = τ = 2.76/6000 = 0.00046 for all g and r. This essentially turns Eq. (5) into a function of model size only. |
iBMA-noprior | Gene expression data only | Iterative BMA without any use of external knowledge. |
LASSO-noprior | Gene expression data only | LASSO without any use of external knowledge. |
LAR-noprior | Gene expression data only | LAR without any use of external knowledge. |