The data consists of case-control pairs. The pair of counts per pair and gene are compared by testing if the proportion of cases is equal to the proportion of the sample using the sizes of library (binomTest). It is assumed given (constant) the total count per pair and gene. The second choice is to use the odds-ratio (statistic or p-value)

oddsratio.paired(x, y, type = "p-value")

Arguments

x

Count matrix

y

Design matrix (nrow(y) == ncol(x)). First column y[,1] is the block and second column y[,2] is the element within the block with values 0 (control) and 1 (case)

type

odds-ratio (stats) or p-value (testing if the odds-ratio is equal to one)

Value

A matrix where each row corresponds with the different expression statistics of the corresponding row of x for the common design given by y