MOFA (version 1.3.1)

getExpectations: getExpectations

Description

Function to extract the expectations from the (variational) posterior distributions of a trained MOFAmodel object.

Usage

getExpectations(object, variable, as.data.frame = FALSE)

Arguments

object

a trained MOFAmodel object.

variable

variable name: 'Z' for factors, 'W' for weights, 'Tau' for noise, 'Y' for pseudodata, 'Theta' for feature-wise spike-and-slab sparsity, 'Alpha' for view and factor-wise ARD sparsity

as.data.frame

logical indicating whether to output the result as a long data frame, default is FALSE.

Value

the output varies depending on the variable of interest:

• "Z": a matrix with dimensions (samples,factors). If as.data.frame is TRUE, a long-formatted data frame with columns (sample,factor,value)

• "W": a list of length (views) where each element is a matrix with dimensions (features,factors). If as.data.frame is TRUE, a long-formatted data frame with columns (view,feature,factor,value)

• "Y": a list of length (views) where each element is a matrix with dimensions (features,samples). If as.data.frame is TRUE, a long-formatted data frame with columns (view,feature,sample,value)

• "Theta": a list of length (views) where each element is a vector of containing the values for each factor

• "Alpha": a list of length (views) where each element is a vector of containing the values for each factor

• "Tau": a list of length (views) where each element is a matrix with dimensions (samples, features)

Details

Technical note: MOFA is a Bayesian model where each variable has a prior distribution and a posterior distribution. In particular, to achieve scalability we used the variational inference framework, thus true posterior distributions are replaced by approximated variational distributions. This function extracts the expectations of the variational distributions, which can be used as final point estimates to analyse the results of the model. The priors and variational distributions of each variable are extensively described in the supplementary methods of the original paper.

Examples

# NOT RUN {
# load a trained MOFAmodel object
filepath <- system.file("extdata", "CLL_model.hdf5", package = "MOFAdata")
MOFAobject <- loadModel(filepath)
# get expectations of Alpha as matrix
getExpectations(MOFAobject, variable="Alpha")
# }