MIMOSA: Fit a MIMOSA Model

In RGLab/MIMOSA: Mixture Models for Single-Cell Assays

Description

This method fits a MIMOSA model to count data stored in an ExpressionSet object.

Usage

MIMOSA(formula, data, ...)

Arguments

formula	describing the features on the lhs and the phenodata on the rhs, supporting extended formula interface with conditioning.
data	an ExpressionSet object with features on rows and samples (labelled with phenoData) on columns.
...	additional arguments

Details

The ExpressionSet should be fully annotated with featureData and phenoData. For ICS data, for example, features would be positive and negative counts for different cytokine producing cell subsets (i.e. IFNg_pos, IFNg_neg) The formula lhs should contain features and the rhs should contain phenotypic variable. See the vignette for an example.

Value

an object of type MIMOSAResult

See Also

MIMOSA-package ConstructMIMOSAExpressionSet MIMOSAResult

Examples

data(ICS)
E<-ConstructMIMOSAExpressionSet(ICS,
  reference=ANTIGEN%in%'negctrl',measure.columns=c('CYTNUM','NSUB'),
  other.annotations=c('CYTOKINE','TCELLSUBSET','ANTIGEN','UID'),
  default.cast.formula=component~UID+ANTIGEN+CYTOKINE+TCELLSUBSET,
  .variables=.(TCELLSUBSET,CYTOKINE,UID),
  featureCols=1,ref.append.replace='_REF')

result<-MIMOSA(NSUB+CYTNUM~UID+TCELLSUBSET+CYTOKINE|ANTIGEN,
    data=E, method='EM',
    subset=RefTreat%in%'Treatment'&ANTIGEN%in%'ENV',
    ref=ANTIGEN%in%'ENV'&RefTreat%in%'Reference')

RGLab/MIMOSA documentation built on Nov. 13, 2020, 5:04 a.m.