Base class for gene model
Y \sim N(X_k\beta + X \gamma, \sigma_Y^2 I_n \times n)
\gamma \sim N(u, \tau^2 I_p \times p)
u = [u_1,...,u_n_c]
u_i \sim N(\mu_0, \sigma^2_u 1
SetupGeneMix(
formula,
data,
X,
tauOff = FALSE,
sigmaMuOff = TRUE,
meanMuOff = FALSE,
nc = NULL,
SVDX = NULL,
testTau = FALSE,
nPC = 0
)
\itemformula- (formula) the covariates formula
\itemdata- (data.frame) for covariates (X_k) these should not include X
\itemX- (n x p) the gene covariates
\itemtauOff- (bool) fix \tau = 0,
\itemsigmaMuOff- (bool) \sigma_u = 0, if sigmaMuOff=FALSE needs to include nc
\itemmeanMuOff- (bool) \mu_0,= 0,
\itemnc- (int) number of chromosones, if null simpler model
\itemSVDX- (obj) singular value decomposition object of X
\itemtestTau- (bool) should we try to turn off tau?
\itemnPC- (int) how many PC to use in fixed effect
Base class for gene model
Y \sim N(X_k\beta + X \gamma, \sigma_Y^2 I_n \times n)
\gamma \sim N(u, \tau^2 I_p \times p)
u = [u_1,...,u_n_c]
u_i \sim N(\mu_0, \sigma^2_u 1