GAP_dicho: GAP_dicho function
In azolling/EBmodules:

This function is a modified version of the GAP statistic for very large datasets.

1 2	GAP_dicho(R, Kmin, Kmax, Nperm = 100, method = "ward.D", bestk_method = "Tib", dicho_f = 2)

`R`	is a correlation matrix
`Kmin`	gives the minimum number of clusters that we want to obtain
`Kmax`	gives the maximum number of clusters that we want to obtain
`Nperm`	is the number of resampling that we want to do
`method`	can only be one of the proposed clustering algorithm. it must return a list with one of the elements named "cluster", which is a vector of length p (number of genes) and containing the id of the cluster for each gene
`bestk_method`	the way to compute the best k as describe in the GAP package
`dicho_f`	the factor for the dichotomy, A small value will lead to several rounds.