som.tree: Build Growing Self-organising Map and Construct a Tree
In JulianSpagnuolo/FACkit: FACS Analysis Tool Kit

Build Growing Self-organising Map and Construct a Tree

1
2
3

som.tree(data, markers, shape = c("rect", "hex"), maxit = 500,
  cores = 1, alpha = 0.9, beta = 0.5, spread = 0.9,
  alg = c("kohonen"), dim = c(NULL, NULL))

`data`	data.frame or matrix containing expression data
`markers`	vector of markers/columns in data to build the SOM with
`shape`	vector determining the shape of SOM to build, either rectangular, "rect", or hexagonal, "hex".
`maxit`	number of iterations the SOM building function will use, default = 500.
`cores`	number of cores to using in distance computation. Default = 1, set higher if you have more than 1 cpu to use.
`alpha`	discount factor for the learning rate during the growing phase of the training. Values should be between 0 and 1. Default = 0.9
`beta`	propagation rate. Determines the rate at which the error of a node, that cannot grow any nodes, is passed on to its neighbours. Suggested values may range between 0 and 1. Default = 0.5
`spread`	numeric value between 0 and 1, controls the rate at which new nodes are created in the growing SOM algorithm, lower values decrease the rate, higher ones increase it. Default is 0.9
`alg`	character vector of length 1. Determines which SOM algorithm to use, choices include either Kohonen SOM, "kohonen" or Growing SOM, "grow" - GrowingSOM is not implemented - awaiting a bug fix and rerelease.
`dim`	integer vector of length 2. Determines dimensions of the SOM in the x and y dimension, respectively. These parameters must be set for the kohonen algorithm. If using the GrowSOM algorithm, defining dim will force the growing SOM to use a gridsize of dimension dim[1]*dim[2]. Default is c(NULL,NULL).