Table 1.

Number of Features in Simulations

Latent Variable	# Features
H	15
G	10
G′	15
Noise in $\mathcal{G}$	1400
S	20
X	15
X′	20
Noise in $\mathcal{X}$	1500

The table summarizes the number of features within each latent variable used in the simulations. These latent variables are displayed in Fig. 7. H describes the subset of features within $\mathcal{G}$ that relate to the outcome but not any features within $\mathcal{X}$, G describes the subset of features within $\mathcal{G}$ belonging to the network, and G^′ describes the subset of features within $\mathcal{G}$ related to some features within $\mathcal{X}$ but not the outcome. S, X, and X^′ describe similar subsets of features within $\mathcal{X}$