generateMixedData: Generate mixed data with or without noise
In ctlab/ClusDec: Clustering approach to solve complete Deconvolution problem

Generates mixed data with or without noise under assumption of linear model

1
2
3

generateMixedData(samples = 40, genes = 12000, cellTypes = 3,
  bias = NULL, pureGenes = 0, noiseDeviation = 0, removeAngles = F,
  cutCoef = 0.85, removeBorders = F, borderShift = 0.25)

`samples`	number of samples
`genes`	number of genes
`cellTypes`	number of cell types
`bias`	if not null bias should be a number in between 0 and 1, one of cell types presented in mix will be more abundunt and one cell type will be less abundunt
`pureGenes`	number of genes which are not noisy
`noiseDeviation`	standart deviation of normally distributed noise value
`removeAngles`	if TRUE there will be no signature genes in the mix (simplex corners will be removed)
`removeBorders`	if TRUE where will be no genes close to simplex border
`borderShift`	number in between 0 and 1, every value in basis is guaranteed to be at least borderShift
`cutCoeff`	corners are removed as points that lie outside of cellTypes-dimensional sphere with radius of cutCoef