This function can be used to learn the networks related to disease prograssion.
network.learn(data_observe,x_cov,X_big_int,X_big_int_cov,
agent,degree=3,len.knots=3,cv=TRUE,nfolds=20,alpha=1)The gene expression matrix.
The vector including covariate values (eg., smoking: 0 and 1).
The matrix based on varying intercepts with the column containing ones. It can be extracted from the output of function `base.construct`.
(X_big_int), The matrix based on varying covariates with the column containing ones. It can be extracted from the output of function `base.construct`.
The imputed disease risk.
The degree in the B-spline base. The default is 3
The number of knots. The default is 3
Use cross-validation or not. The default is TRUE. If cv=FALSE, then BIC method will be applied.
The number of folds used in cross-validation. Only workable when cv=TRUE.
0 or 1 values: 1: LASSO; 0: ridge regression.
The list of network properties: self-node size for baseline, self-node size for covariate effect, overall self-node size, gene interaction effects from the baseline, gene interaction effects from the covariate, and overall gene interaction effects