simulateSparseDataSet: Simulates a SparseDataSet

In mikelove/SparseData: Efficient Manipulation of Sparse Experiment Data

Description

Given a number of features n, and the number of samples per condition, this returns a SparseDataSet object with random data. The samples in each condition follow a negative binomial distribution.

The means for the distribution are a sum of a global and sample specific vector. The nonzero elements for these are generated by a gamma distribution, and the proportion of nonzero are controlled by nzg and nzs. The resulting matrix will have then at most (nzg + nzs * nconditions) nonzero elements.

Usage

simulateSparseDataSet(n, samples.per.condition,nzg=.1,nzs=.1)

Arguments

n	an integer, the number of features to generate
samples.per.condition	a vector of integers, the number of samples per condition
nzg	average number of nonzero for the global mu
nzs	average number of nonzero for the sample mus

Value

a SparseDataSet object

Author(s)

Michael Love

Examples

sds <- simulateSparseDataSet(100, c(5,5))

mikelove/SparseData documentation built on May 22, 2019, 10:52 p.m.