edgenet {netReg} | R Documentation |
Fit a graph-regularized linear regression model using edge-penalization. The coefficients are computed using graph-prior knowledge in the form of one/two affinity matrices. Graph-regularization is an extension to previously introduced regularization techniques, such as the LASSO. For that reason we are also using coordinate descent for minimization of the objective function of the linear model.
edgenet(X, Y, G.X = NULL, G.Y = NULL, lambda = 1, psigx = 1, psigy = 1, thresh = 1e-05, maxit = 1e+05, family = c("gaussian"))
X |
input matrix, of dimension ( |
Y |
output matrix, of dimension ( |
G.X |
non-negativ affinity matrix for |
G.Y |
non-negativ affinity matrix for |
lambda |
|
psigx |
|
psigy |
|
thresh |
|
maxit |
maximum number of iterations for coordinate descent
( |
family |
family of response, e.g. gaussian |
An object of class edgenet
coefficients |
the estimated ( |
intercept |
the estimated ( |
call |
the call that produced the object |
family |
the family of the response |
Dirmeier, Simon and Fuchs, Christiane and Mueller, Nikola S and Theis,
Fabian J (2018),
netReg: Network-regularized linear models for biological association
studies.
Bioinformatics
Friedman J., Hastie T., Hoefling H. and Tibshirani R. (2007),
Pathwise coordinate optimization.
The Annals of Applied Statistics
Friedman J., Hastie T. and Tibshirani R. (2010),
Regularization Paths for Generalized Linear Models via
Coordinate Descent.
Journal of Statistical Software
Fu W. J. (1998), Penalized Regression: The Bridge Versus the Lasso.
Journal of Computational and Graphical Statistics
Cheng W. and Wang W. (2014), Graph-regularized dual Lasso for robust
eQTL mapping.
Bioinformatics
X <- matrix(rnorm(100*10), 100, 10) b <- rnorm(10) G.X <- matrix(rpois(100,1), 10) G.X <- t(G.X) + G.X diag(G.X) <- 0 # fit a Gaussian model Y <- X%*%b + rnorm(100) fit <- edgenet(X=X, Y=Y, G.X=G.X, family="gaussian")