MIDTList: Infrastructure for Multiple Imputation and Multi-omics...

In GonzalezIgnacio/missRows: Handling Missing Individuals in Multi-Omics Data Integration

Description

MIDTList is an S4 class that extends the class MultiAssayExperiment by providing the infrastructure (slots) to store the input data, intermediate calculations and results of a multiple imputation approach.

Objects from the Class

Objects can be created by calls of the form:

MIDTList(..., colData=NULL, strata=NULL, assayNames=NULL)

new("MIDTList", ..., colData=NULL, strata=NULL, assayNames=NULL)

Arguments

...

arguments passed to '...' can be:

1) data tables with missing individuals. Two or more objects which can be interpreted as matrices (or data frames). Data tables passed as arguments in ... must be arranged in variables (rows) x individuals (columns), with individual names matching row names of colData.

2) a list containing two or more data tables with missing individuals. Data tables (matrices or data frames) within list must be arranged in variables (rows) x individuals (columns), with individual names matching row names of colData.

3) an object of class MultiAssayExperiment. In this case colData and assayNames arguments are ignored.

colData

a DataFrame giving the characteristics for all individuals (biological units). The row names of colData must contain individual identifiers.

assayNames

optional. A character vector giving the name for each table.

strata

a character indicating the column of colData to be used as strata in the construction of MIDTList.

Details

To facilitate programming pipelines, NULL values are input for compromise, configurations, imputedIndv and MIparam slots, in which case the default value is used as if the argument had been missing. These slots will be updated after multiple imputation (MIMFA) approach.

Slots

ExperimentList

an ExperimentList class object for each assay dataset.

colData

a DataFrame of all clinical/specimen data available across experiments.

sampleMap

a DataFrame of translatable identifiers of samples (individuals) and participants.

metadata

additional data describing the MultiAssayExperiment object.

drops

a metadata list of dropped information.

strata:

a numeric value or character. The column of colData to be used as strata in MIMFA.

missingIndv:

a list containing character vectors with the name of the missing individuals per table.

compromise:

the compromise configuration, a matrix with the individuals coordinates as returned by STATIS function.

configurations:

a list containing the individuals coordinates for each imputed dataset as returned by MIMFA function.

imputedIndv:

a list containing the imputed individuals for each data table as returned by imputeDataMFA function.

MIparam:

a list containing the parameters used in the MIMFA function.

Extends

Class MultiAssayExperiment, directly.

Methods

initialize

signature(.Object = "MIDTList"): See 'Objects from the Class' section for description.

Class-specific methods return the corresponding objects:

strata

signature(object="MIDTList"): Return factor of strata giving the stratum for each individual.

missingIndv

signature(object="MIDTList"): Return list containing character vectors with the name of the missing individuals per table.

compromise

signature(object="MIDTList"): Return matrix with the individuals coordinates as returned by STATIS function.

configurations

signature(object="MIDTList", M="all"): Get all configurations. If M is a positive integer, the Mth configuration is returned.

imputedIndv

signature(object="MIDTList"): Return list containing the imputed individuals for each data table.

MIparam

signature(object="MIDTList"): Return list containing the parameters used in the MIMFA function.

Standard generic methods:

show

signature(object="MIDTList"): Informatively display object contents.

See MultiAssayExperiment-class for generic methods associated to the MultiAssayExperiment class.

Author(s)

Ignacio González

See Also

MultiAssayExperiment-class, MultiAssayExperiment-methods

Examples

#-- load data
data(NCI60)

#-- MIDTList object from separate data tables
table1 <- NCI60$dataTables$trans
table2 <- NCI60$dataTables$prote
colData <- NCI60$dataTables$cell.line

midt <- MIDTList(table1, table2, colData=colData,
                assayNames=c("transcrip", "proteome"))
midt

#-- MIDTList object from a list
tablesList <- NCI60$dataTables[1:2]
colData <- NCI60$dataTables$cell.line

midt <- MIDTList(tablesList, colData=colData)
midt

#-- MIDTList object directly from a 'MultiAssayExperiment'
midt <- MIDTList(NCI60$mae)
midt

GonzalezIgnacio/missRows documentation built on Jan. 16, 2020, 4:11 a.m.