permutation: Permute group labels

In MetaLonDA: Metagenomics Longitudinal Differential Abundance Method

Description

Permutes the group label of the samples in order to construct the AR empirical distibution

Usage

permutation(
  perm.dat,
  n.perm = 500,
  method = "nbinomial",
  points,
  lev,
  parall = FALSE
)

Arguments

perm.dat	dataframe has the Count, Group, ID, Time
n.perm	number of permutations
method	The fitting method (negative binomial, LOWESS)
points	The points at which the prediction should happen
lev	the two level's name
parall	boolean to indicate whether to use multicore.

Value

returns the fitted model for all the permutations

References

Ahmed Metwally ([e-mail])

Examples

data(metalonda_test_data)
n.sample = 5
n.timepoints = 10
n.perm = 3
n.group = 2
Group = factor(c(rep(0, n.sample*n.timepoints), rep(1, n.sample*n.timepoints)))
Time = rep(rep(1:n.timepoints, times = n.sample), 2)
ID = factor(rep(1:(2*n.sample), each = n.timepoints))
points = seq(1, 10, length.out = 10)
aggregate.df = data.frame(Count = metalonda_test_data[1,], Time = Time, Group = Group, ID = ID)
prm = permutation(aggregate.df, n.perm = 3, method = "nbinomial", points)

Example output

# of subjects =  10 
Permutation =  1 
Permutation =  2 
Permutation =  3 

MetaLonDA documentation built on Dec. 18, 2019, 9:38 a.m.