R/AllClasses.R

In slalom: Factorial Latent Variable Modeling of Single-Cell RNA-Seq Data

# Define the SlalomModel class for the package

#' The "Slalom Model" (Rcpp_SlalomModel)  class
#'
#' S4 class and the main class used by slalom to hold model data and results.
#' SingleCellExperiment extends the Bioconductor SummarizedExperiment class.
#'
#' This class is initialized from a matrix of expression values and a collection
#' of genesets in a \code{GeneSetCollection} object from the GSEABase package.
#'
#' Methods that operate on SingleCellExperiment objects constitute the basic scater workflow.
#'
#' @section Slots:
#'  \describe{
#'    \item{\code{.xData}:}{Environment enabling access to the C++-level
#'    SlalomModel object.}
#'}
#'
#' @details
#'
#' @name Rcpp_SlalomModel
#' @rdname Rcpp_SlalomModel
#' @aliases Rcpp_SlalomModel-class
#' @exportClass Rcpp_SlalomModel
setClass("Rcpp_SlalomModel",
        slots = c(.xData = "environment")
)

# slots = c(K = "numeric",
#           N = "numeric",
#           G = "numeric",
#           nScale = "numeric",
#           nAnnotated = "numeric",
#           nHidden = "numeric",
#           nKnown = "numeric",
#           noiseModel = "character",
#           iUnannotatedDense = "integer",
#           iUnannotatedSparse = "integer",
#           nIterations = "integer",
#           minIterations = "integer",
#           iterationCount = "integer",
#           forceIterations = "logical",
#           tolerance = "numeric",
#           shuffle = "logical",
#           nOn = "numeric",
#           onF = "numeric",
#           doUpdate = "logical",
#           learnPi = "logical",
#           dropFactors = "logical",
#           termNames = "character",
#           geneNames = "character",
#           cellNames = "character",
#           I = "matrix",
#           Known = "matrix",
#           Y = "matrix",
#           pseudo_Y = "matrix",
#           YY = "matrix",
#           ## alpha
#           alpha_pa = "numeric",
#           alpha_pb = "numeric",
#           alpha_a = "numeric",
#           alpha_b = "numeric",
#           alpha_E1 = "numeric",
#           alpha_lnE1 = "numeric",
#           ## epsilon
#           epsilon_pa = "numeric",
#           epsilon_pb = "numeric",
#           epsilon_a = "numeric",
#           epsilon_b = "numeric",
#           epsilon_E1 = "numeric",
#           epsilon_lnE1 = "numeric",
#           epsilon_diagSigmaS = "numeric",
#           ## X
#           X_E1 = "matrix",
#           X_diagSigmaS = "numeric",
#           X_init = "matrix",
#           ## W
#           W_E1 = "matrix",
#           W_sigma2 = "matrix",
#           W_E2diag = "matrix",
#           W_gamma0 = "matrix",
#           W_gamma1 = "matrix",
#           ## Z
#           Z_E1 = "matrix",
#           Z_init = "matrix",
#           ## Pi
#           Pi_a = "numeric",
#           Pi_pa = "numeric",
#           Pi_pb = "numeric",
#           Pi_b = "numeric",
#           Pi_E1 = "numeric"
# )

#'  \describe{
#'    \item{\code{K}:}{Scalar of class \code{"numeric"}, indicating total number
#'     of factors.}
#'    \item{\code{N}:}{Scalar of class \code{"numeric"}, indicating number of
#'    cells.}
#'    \item{\code{G}:}{Scalar of class \code{"numeric"}, indicating number of
#'    genes.}
#'    \item{\code{nScale}:}{Scalar of class \code{"numeric"}, relative number of
#'     cells to which the observed annotation data (gene sets) are scaled.}
#'    \item{\code{nAnnotated}:}{Scalar of class \code{"numeric"}, number of
#'    factors relating to annotated gene sets to model.}
#'    \item{\code{nHidden}:}{Scalar of class \code{"numeric"}, number of hidden
#'    (latent) factors to model.}
#'    \item{\code{nKnown}:}{Scalar of class \code{"numeric"}, number of known
#'    factors (covariates).}
#'    \item{\code{noiseModel}:}{\code{"character"} scalar defining noise model
#'    used by the model (default: "gauss" for Gaussian noise model.}
#'    \item{\code{iUnannotatedDense}:}{\code{"integer"} vector giving indices for
#'    dense unannotated (hidden) factors.}
#'    \item{\code{iUnannotatedSparse}:}{\code{"integer"} vector giving indices for
#'    sparse unannotated (hidden) factors.}
#'    \item{\code{nOn}:}{Vector of class \code{"numeric"}, number of genes that
#'    are "on" for each factor (as annotated).}
#'    \item{\code{termNames}:}{\code{"character"} vector giving names for the gene
#'    sets and factors.}
#'    \item{\code{geneNames}:}{\code{"character"} vector giving gene names.}
#'    \item{\code{cellNames}:}{\code{"character"} vector giving cell names.}
#'    \item{\code{X}:}{\code{"list"} of values and parameters for factor states.}
#'    \item{\code{W}:}{\code{"list"} of values and parameters for factor weights.}
#'    \item{\code{Z}:}{\code{"list"} of values and parameters for activation
#'    variable of whether factor k has a regulatory effect on gene g.}
#'    \item{\code{alpha}:}{\code{"list"} of values and parameters for factor
#'    precisions.}
#'    \item{\code{epsilon}:}{\code{"list"} of values and parameters for residual
#'    precisions.}
#'    \item{\code{pi}:}{\code{"matrix"} of size G x K with each entry being the
#'    prior probability for a gene g being active for factor k.}
#'    \item{\code{I}:}{\code{"matrix"} of size G x K of observed annotation data
#'     with each entry being the indicator that gene g is annotated to factor k.}
#'    \item{\code{Known}:}{design \code{"matrix"} defining covariates to fit in
#'    the model ("known factors").}
#'    \item{\code{Y}:}{\code{"matrix"} of size N x G with each entry being the
#'    observed expression value (normalized, log2-scale) for gene g in cell n.}
#'    \item{\code{pseudo_Y}:}{\code{"matrix"} of size N x G with each entry
#'    being the pseudoexpression value (normalized, log2-scale) for gene g in
#'    cell n.}
#'}