Filter for Expressed Genes — Expressed • myTAI

This function takes an ExpressionSet object and removes genes from the gene expression matrix that have an expression level below, above, or below AND above a defined cut.off value. Hence, this function allows to remove genes that have been defined as not expressed or outliers and returns an ExpressionSet retaining only expressed genes.

Usage

Expressed(
  ExpressionSet,
  cut.off,
  method = "const",
  comparison = "below",
  n = NULL
)

Arguments

ExpressionSet: a standard PhyloExpressionSet or DivergenceExpressionSet object.
cut.off: a numeric value specifying the expression cut-off to define genes as not expressed (comparison = "below") , outliers (comparison = "above"), or both (comparison = "both"). See comparison for details. In case comparison = "both", the cut.off argument must be a two dimensional vector defining the lower cut.off value at the first position and the upper cut.off value at the second position.
method: a method defining how to treat gene expression values in multiple stages. The corresponding method that is chosen allows to control the stage-wise fulfillment of the threshold criteria. Options are "const", "min-set", and "n-set".
comparison: a character string specifying whether genes having expression levels below, above, or below AND above (both) the cut.off value should be excluded from the dataset. In case comparison = "both" is chosen, the cut.off argument must be a two dimensional vector defining the lower cut.off value at the first position and the upper cut.off value at the second position.
n: a numeric value for method = "n-set".

Details

This filter function allows users to remove genes from the ExpressionSet object that undercut or exceed a certain expression level cut.off.

Following extraction criteria are implemented in this function:

const: all genes that have at least one stage that undercuts or exceeds the expression cut.off will be excluded from the ExpressionSet. Hence, for a 7 stage ExpressionSet genes passing the expression level cut.off in 6 stages will be retained in the ExpressionSet.
min-set: genes passing the expression level cut.off in ceiling(n/2) stages will be retained in the ExpressionSet, where n is the number of stages in the ExpressionSet.
n-set: genes passing the expression level cut.off in n stages will be retained in the ExpressionSet. Here, the argument n needs to be specified.

Author

Hajk-Georg Drost

Examples

data(PhyloExpressionSetExample)

# remove genes that have an expression level below 8000 
# in at least one developmental stage
FilterConst <- Expressed(ExpressionSet = PhyloExpressionSetExample, 
                         cut.off       = 8000, 
                         method        = "const",
                         comparison    = "below")
                              
dim(FilterConst) # check number of retained genes
#> [1] 449   9

# remove genes that have an expression level below 8000 
# in at least 3 developmental stages 
# (in this case: ceiling(7/2) = 4 stages fulfilling the cut-off criteria)
FilterMinSet <- Expressed(ExpressionSet = PhyloExpressionSetExample, 
                          cut.off       = 8000, 
                          method        = "min-set",
                          comparison    = "below")
                               
dim(FilterMinSet) # check number of retained genes
#> [1] 830   9

# remove genes that have an expression level below 8000 
# in at least 5 developmental stages (in this case: n = 2 stages fulfilling the criteria)
FilterNSet <- Expressed(ExpressionSet = PhyloExpressionSetExample, 
                        cut.off       = 8000, 
                        method        = "n-set",
                        comparison    = "below",
                        n             = 2)
                               
dim(FilterMinSet) # check number of retained genes
#> [1] 830   9



# remove expression levels that exceed the cut.off criteria
FilterMinSet <- Expressed(ExpressionSet = PhyloExpressionSetExample, 
                          cut.off       = 12000, 
                          method        = "min-set",
                          comparison    = "above")
                               
dim(FilterMinSet) # check number of retained genes
#> [1] 24456     9


# remove expression levels that undercut AND exceed the cut.off criteria
FilterMinSet <- Expressed(ExpressionSet = PhyloExpressionSetExample, 
                          cut.off       = c(8000,12000), 
                          method        = "min-set",
                          comparison    = "both")
                               
dim(FilterMinSet) # check number of retained genes
#> [1] 111   9