Fit the loading matrix Z using sparse regression of prior information U

For each column of a target matrix Z using pathway or prior annotation priorMat. It performs regularization selection using cross-validation and can apply either Supports continuous or binary response models. Relaxed refitting is supported for final coefficient estimation.

Usage

solveU(
  Z,
  Chat = NULL,
  priorMat,
  penalty.factor,
  pathwaySelection = c("fast", "complete"),
  alpha = 0.9,
  maxPath = 10,
  nfolds = 5,
  useSE = FALSE,
  top = NULL,
  binary = FALSE,
  nlambda = 20,
  scale = TRUE,
  refit = TRUE,
  Uprev = NULL,
  useAUC = TRUE,
  intercept = TRUE,
  ...
)

Arguments

Z

A numeric matrix with features (rows) and samples (columns).

Chat

(Optional) Precomputed pseudo-inverse of priorMat; if NULL, it is calculated using ridge regularization.

priorMat

A numeric matrix with prior information (features x pathways).

penalty.factor

Optional penalty weights for features in priorMat.

pathwaySelection

Method to select candidate pathways: "fast" (default) or "complete".

alpha

Elastic net mixing parameter (0 = ridge, 1 = lasso). Default is 0.9.

maxPath

Maximum number of pathways/features selected per column. Default is 10.

nfolds

Number of cross-validation folds. Default is 5.

useSE

Whether to use the 1-standard-error rule for lambda selection. Default is FALSE.

top

If set, sets to 0 all but the top entries of Z per column before fitting.

binary

If TRUE, fits a binomial model (e.g., classification) to Z>0. Can be used in combination with top. Default is FALSE.

nlambda

Number of lambda values for glmnet. Default is 20.

scale

Whether to standardize predictors in glmnet. Default is TRUE.

refit

Whether to perform relaxed refitting using selected predictors. Default is TRUE.

Uprev

(Optional) Previous U matrix to reuse. In this mode only the columns of U that are all zero are estimated. Used internally in CLAMP.

useAUC

Logical; whether to compute pathway-LV associations using AUC (default TRUE) instead of OLS.

intercept

Logical; whether to include an intercept term in glmnet models. Default is TRUE.

...

Additional arguments passed to glmnet() or cv.glmnet().

Value

A list with one element:

U

A matrix of loadings (features x components). Columns are named LV1, LV2, ...

Examples

set.seed(123)
genes <- paste0("G", 1:200)
lvs <- paste0("LV", 1:4)
paths <- paste0("Path", 1:60)

Z <- matrix(rnorm(200 * 4), nrow = 200, dimnames = list(genes, lvs))

priorMat <- matrix(rbinom(200 * 60, 1, 0.07),
    nrow = 200, dimnames = list(genes, paths)
)

fit1 <- solveU(
    Z = Z,
    priorMat = priorMat,
    pathwaySelection = "fast",
    alpha = 0.9,
    maxPath = 10,
    nfolds = 5,
    binary = FALSE,
    refit = TRUE
)
#> Warning: Passing 'dfmax' to glmnet() is deprecated. Use control = list(dfmax = ...) instead.