Table 1 Simulation settings.
From: An exploratory data analysis method to reveal modular latent structures in high-throughput data
Parameters | Values | Â |
|---|---|---|
Modular factor model | Â | Â |
Type of hidden factors | Gaussian, mixed | Â |
Number of samples | 100 | Â |
Number of modules | 5 | 10 |
Number of genes per module | 200 | 100 |
Maximum number of factors per module | 4 | 3 |
Minimum number of factors per module | 1 | Â |
% non-zero loadings within module | 40%, 70%, 100% | Â |
Number of pure noise genes | 200, 1000 | Â |
Signal to noise ratio | 0.5, 1, 2 | Â |
Global sparse factor model | Â | Â |
Type of hidden factors | Gaussian, mixed | Â |
Number of samples | 100 | Â |
Number of genes | 2500 # | Â |
Number of factors | 20 | Â |
Average number of factors governing each gene | 0.5, 1, 2, 5, 10, 20 | Â |
Signal to noise ratio | 0.5, 1, 2 | Â |