MOFA (version 1.3.1)

getDefaultModelOptions: Get default model options

Description

Function to obtain the default model options.

Usage

getDefaultModelOptions(object)

Arguments

object

an untrained MOFAmodel object

Value

Returns a list with the default model options, which have to be passed as an argument to prepareMOFA

Details

The model options are the following:

• likelihood: character vector with data likelihoods per view: 'gaussian' for continuous data, 'bernoulli' for binary data and 'poisson' for count data. By default, they are guessed internally.

• numFactors: numeric value indicating the initial number of factors. If you want to learn the number of factors automatically we recommend setting this to a large value, default is 25.

• sparsity: logical indicating whether to use sparsity. This is always recommended, as it will make the loadings more interpretable. Default is TRUE.

Examples

# NOT RUN {
# load data
data("CLL_data", package = "MOFAdata")
#create a MOFAmodel object
MOFAobject <- createMOFAobject(CLL_data)
# set model options
ModelOptions <- getDefaultModelOptions(MOFAobject)
ModelOptions
# }