optim_clusters_coord: Select optimal number of clusters using the Gap statistic
In sebastianduchene/clockstarG: ClockstaR-G: ClockstaR for genomic data sets

Description Usage Arguments Value Author(s) References

View source: R/optim_clusters_coord.R

optim_clusters_coord obtains the optimal number of clusters, the cluster assignment, and other cluster statistics for the optimal number of clusters, obtained with the Gap statistic.

optim_clusters_coord(coord_mat, n_clusters = 2, kmax, b_reps = 100, out_cluster_id = "opt_clus_id.txt", out_cluster_info = "opt_clusinfo_sbsd.txt", out_gap_stats = "gap_stats.txt", plot_clustering = F)

`coord_mat`	This is the matrix of the scaled branch lengths of the gene trees. It can be obtained with get_scaled_brs.
`n_clusters`	Number of computing clusters. This is the number of CPUs to use. Do not confuse with the number of clocks or pacemakers.
`kmax`	Maximum number of clusters to test. This is k for 1<k<N
`b_reps`	Bootstrap replicates
`out_cluster_id`	Name of the file to save the cluster assignment for each gene.
`out_cluster_info`	Name of the file to save the cluster information for each gene. This includes issolation, mean dissimilarity, and cluster size.
`out_gap_stats`	Name of file to save the Gap statistic for values of k.
`plot_clustering`	boolean. Should ClockstaR-G plot the Gap statistics.

`optimal_k`	The optimal value for k.
`cluster_info`	matrix with information for each of the clusters (referred to as pacemakers in the publication)
`cluster_id`	matrix with the clustering assignment for each gene
`gap_statistics`	matrix with the Gap statistics
`alt_gap`	difference between sucessive mean Gap statistics