Calculate MAD scores as per OmicSoft — mad

Folllows description at https://wiki.arrayserver.com/wiki/index.php?title=CorrelationQC.pdf.

Usage

mad_score(matrix, sample_sheet = NULL, groupby = NULL, outlier_threshold = -5)

Arguments

matrix: Matrix with samples by column
sample_sheet: Sample sheet with samples by row
groupby: Sample sheet column that can be used to group samples
outlier_threshold: Value below which points should be flagged as outliers, conventionally -5

Value

mad_score A data frame with columns for group name, MAD score and outlier status. A threshold of < -5 usually indicates outliers

Details

Not currently deployed anywhere in shinyngs, but a potential way of flagging outliers for investigation

Examples

mat <- matrix(rnorm(60), nrow = 6, ncol = 10,
  dimnames = list(paste0("gene", 1:6), paste0("s", 1:10)))
sample_sheet <- data.frame(
  condition = rep(c("treated", "control"), each = 5),
  row.names = colnames(mat)
)
mad_score(mat, sample_sheet, groupby = "condition")
#>       group        mad outlier
#> s6  control  0.6744908   FALSE
#> s7  control  0.1248481   FALSE
#> s8  control -7.3259914    TRUE
#> s9  control -1.0314661   FALSE
#> s10 control  0.0000000   FALSE
#> s1  treated  0.0000000   FALSE
#> s2  treated -0.2072914   FALSE
#> s3  treated  0.6744908   FALSE
#> s4  treated  0.9354219   FALSE
#> s5  treated -2.2211295   FALSE