MOFA (version 1.3.1)

createMOFAobject: Initialize a MOFA object

Description

Method to initialize a MOFAmodel object with a multi-omics data set.

Usage

createMOFAobject(data)

Arguments

data

either a MultiAssayExperiment or a list of matrices with features as rows and samples as columns.

Value

Returns an untrained MOFAmodel object. Next step is to define the training, model and data processing options (see prepareMOFA)

Details

If the multi-omics data is provided as a list of matrices, please make sure that features are stored as rows and samples are stored as columns. If the matrices have sample names, we will use them to match the different matrices, filling the corresponding missing values. If matrices have no column names, all matrices must have the same number of columns, and you are responsible for filling any missing values.

Examples

# NOT RUN {
# Option 1: Create a MOFAobject from a list of matrices, features in rows and samples in columns.
data("CLL_data", package = "MOFAdata")
MOFAobject <- createMOFAobject(CLL_data)

# Option 2: Create a MOFAobject from a MultiAssayExperiment
library(MultiAssayExperiment)
data("CLL_data", package = "MOFAdata")
data("CLL_covariates", package = "MOFAdata")
mae_CLL <- MultiAssayExperiment(experiments = CLL_data, colData = CLL_covariates)
MOFAobject <- createMOFAobject(mae_CLL)

# next, this object can be passed to prepareMOFA and runMOFA
# (training in runMOFA can take some time):
# }
# NOT RUN {
# MOFAobject <- prepareMOFA(MOFAobject)
# MOFAobject <- runMOFA(MOFAobject)
# }