R/classes.R

In genArise: Microarray Analysis tool

require(methods) # In case R version is earlier than 1.7.0

# Class to define a microarray
# Contains 2 attributes:
# name: The name of the experiment
# spotData: The data of the experiment. Cy3, Cy5, BgCy3, BgCy5, Id
setClass("Spot",
	  representation(name ="character", spotData ="list"), where = .GlobalEnv)

# Some checking to get a correct object
setValidity("Spot",function(object){

	spotName <- as.character(attr(object, "name"))
	spotData <- attr(object, "spotData")

	Cy3 <- spotData$Cy3
	Cy5 <-spotData$Cy5
	BgCy3 <- spotData$BgCy3
	BgCy5 <-spotData$BgCy5
	Id <- spotData$Id
        Symdesc <- spotData$Symdesc
        
	if(is.null(Cy3) || is.null(Cy5)) return("Element R or G missing")
	if(!is.numeric(Cy3) || !is.numeric(Cy5)) return("R or G contain non-numeric elements")
	if(length(dim(Cy3)) > 2) return("R and G have more than two dimensions")
})

# Class to define the Z-score for the data set.
# Contains 3 attributes:
# name: The name of the experiment
# dataSet: The data of the experiment. Cy3, Cy5, Id
# type: The analysis wll be done R vs I or M vs A
setClass("DataSet", representation(name = "character", dataSets = "list", type = "character"), where = .GlobalEnv)

# Some checking to get a correct object 
setValidity("DataSet", function(object){
  spot.name <- attr(object, "name")
  dataSets <- attr(object, "dataSets")
  type <- attr(object, "type")
  Cy3 <-  dataSets$Cy3
  Cy5 <-  dataSets$Cy5
  Id  <-  dataSets$Id
  Symdesc <- dataSets$Symdesc
  Zscore <- dataSets$Zscore
  if(!(type == "ri" || type == "ma")) return ("Type argument must be ri or ma")
  if(is.null(Cy3) || is.null(Cy5) || is.null(Id) || is.null(Zscore)) return("Some element is missing")
  if(!is.numeric(Cy3) || !is.numeric(Cy5) || !is.numeric(Zscore)) return("There is a non-numeric element")
})