sisal: SISAL algorithm

In ctlab/LinSeed: Linear Subpsace identification to solve complete gene expression deconvolution problem

Description

Sisal alogorithm for simplex endpoints identification

Usage

sisal(Y, p, iters = 80, tau = 1, mu = p * 1000/ncol(Y),
  spherize = F, tol = 0.01, m0 = NULL, verbose = F,
  returnPlot = T, nonNeg = F)

Arguments

Y	gene expression matrix
p	number of endpoints
iters	number of iterations to perform
tau	noise points penalty coefficient
mu	regularization
spherize	spherize or not
tol	numeric tolerance
m0	starting points for SISAL algorithm, default points are getting from VCA
verbose	verbosity
returnPlot	logical, is it needed to return dataframe or not
nonNeg	logical, force simplex corners to non-negative space

Details

Implementation of method

ctlab/LinSeed documentation built on Aug. 9, 2019, 4:33 p.m.