From: Sparse regression models for unraveling group and individual associations in eQTL mapping
Symbols | Description |
---|---|
K | Number of SNPs |
N | Number of genes |
H | Number of samples |
M | Number of group-wise associations |
x | Random variables of K SNPs |
z | Random variables of N genes |
y | Latent variables to model group-wise associaiton |
\(\mathbf {X}\in {\mathbb {R}}^{K \times H}\) | SNP matrix data |
\(\mathbf {Z} \in {\mathbb {R}}^{N \times H}\) | Gene expression matrix data |
\(\mathbf {A} \in \mathbb {R}^{M\times K}\) | Group-wise association coefficient matrix between x and y |
\(\mathbf {B} \in \mathbb {R}^{N\times M}\) | Group-wise association coefficient matrix between y and z |
\(\mathbf {C} \in \mathbb {R}^{N\times K}\) | Individual association coefficient matrix between x and y |
α,β,γ,ρ | Regularization parameters |
\(\mathbf {R} \in \mathbb {R}^{N\times K}\) | Indicator matrix showing which elements in C can be nonzero |