tweeDExact: Exact test for differences between two Poisson-Tweedie groups

Description

Carry out an exact test for differences between two Poisson-Tweedie populations.

Usage

tweeDExact(counts, group, tol = 1e-15, mc.cores = 1)
exactTestPT(counts, group, tol = 1e-15, threshold = 150e3)

Arguments

counts

The RNA-seq counts. An object of type 'matrix' or 'data.frame' for 'tweeDExact', or an object of type 'vector' for 'exactTest'.

group

vector giving the experimental group/condition for each sample/library.

tol

Tolerance for the Poisson-Tweedie probability computations. The probabilities under the 'tol' value will automatically considered as 0.

threshold

an integer (default is 50e3). If the sum of all counts in a certain gene excedes this value 'testPoissonTweedie' will be called instead of 'exactTest'. Larger values will result in a longer computing time.

mc.cores

number of cpu cores to be used. This option is only available when the 'multicore' package is installed and loaded first. In such a case, if the default value of 'mc.cores=1' is not changed, all available cores will be used.

Value

'exactTest' returns the p-value resulting from the exact test between two different Poisson-Tweedie populations, as well as the method that was used to compute it.'tweeDExact' returns a 'data.frame'. Each row corresponds to a gene and it contains the following information:- In the first columns the mean of counts in each of the subgroups.- In the third column the p-value of the test for differential expression between the two subgroups.- In the fourth column the p-value corrected for multiple comparisons using the Benjamini-Hochberg FDR procedure.- In the last (fifth) column the method that was used to compute the p-value.

Details

'exactTest' performs the exact test for a vector of counts.

'tweeDExact' performs the test for a whole 'data.frame'. The P-values are then corrected using the Benjamini and Hochberg method.

References

P. Hougaard, M.L. Ting Lee, and G.A. Whitmore (1997). Analysis of overdispersed count data by mixtures of poisson variables and poisson processes. Biometrics 53, pages 1225-1238.

Examples

Run this code

counts <- matrix(rPT(n = 1000, a = 0.5, mu = 10, D = 5), ncol = 40)

tweeDExact(counts, group = rep(c(1,2),20))

Run the code above in your browser using DataLab