netboost: Netboost clustering.
In netboost: Network Analysis Supported by Boosting

Description Usage Arguments Value Examples

The Netboost clustering is performed in three subsequent steps. First, a filter of important edges in the network is calculated. Next, pairwise distances are calculated. Last, clustering is performed. For details see Schlosser et al. doi...

netboost(datan = NULL, stepno = 20L, filter_method = c("boosting",
  "skip", "kendall", "spearman", "pearson"), until = 0L,
  progress = 1000L, mode = 2L, soft_power = NULL,
  max_singleton = ncol(datan), qc_plot = TRUE, min_cluster_size = 2L,
  ME_diss_thres = 0.25, n_pc = 1, robust_PCs = FALSE,
  nb_min_varExpl = 0.5, cores = as.integer(getOption("mc.cores", 2)),
  scale = TRUE, method = c("pearson", "kendall", "spearman"),
  verbose = getOption("verbose"))

`datan`	Data frame were rows correspond to samples and columns to features.
`stepno`	Integer amount of boosting steps applied in the filtering step
`filter_method`	The following filtering methods are supported: "boosting" (non-zero coefficients in likelihood based boosting), "skip" (no filter), "kendall" (stats::cor.test), "spearman" (stats::cor.test), "pearson" (stats::cor.test)
`until`	Stop at index/column (if 0: iterate through all columns). For testing purposes in large datasets.
`progress`	Integer. If > 0, print progress after every X steps ( Progress might not be reported completely accurate due to parallel execution)
`mode`	Integer. Mode (0: x86, 1: FMA, 2: AVX). Features are only available if compiled accordingly and available on the hardware.
`soft_power`	Integer. Exponent of the transformation. Set automatically based on the scale free topology criterion if unspecified.
`max_singleton`	Integer. The maximal singleton in the clustering. Usually equals the number of features.
`qc_plot`	Logical. Should plots be created?
`min_cluster_size`	Integer. The minimum number of features in one module.
`ME_diss_thres`	Numeric. Module Eigengene Dissimilarity Threshold for merging close modules.
`n_pc`	Number of principal components and variance explained entries to be calculated. The number of returned variance explained entries is currently ‘min(n_pc,10)’. If given ‘n_pc’ is greater than 10, a warning is issued.
`robust_PCs`	Should PCA be calculated on ranked data (Spearman PCA)? Rotations will not correspond to original data if this is applied.
`nb_min_varExpl`	Minimum proportion of variance explained for returned module eigengenes. The number of PCs is capped at n_pc.
`cores`	Integer. Amount of CPU cores used (<=1 : sequential)
`scale`	Logical. Should data be scaled and centered?
`method`	A character string specifying the method to be used for correlation coefficients.
`verbose`	Additional diagnostic messages.

dendros A list of dendrograms. For each fully separate part of the network an individual dendrogram.

names A vector of feature names.

colors A vector of numeric color coding in matching order of names and module eigengene names (color = 3 -> variable in ME3).

MEs Aggregated module measures (Module eigengenes).

var_explained Proportion of variance explained per module eigengene per principal component (max n_pc principal components are listed).

rotation Matrix of variable loadings divided by their singular values. datan

filter Filter-Matrix as generated by the nb_filter function.

data('tcga_aml_meth_rna_chr18',  package='netboost')
results <- netboost(datan=tcga_aml_meth_rna_chr18, stepno=20L,
   soft_power=3L, min_cluster_size=10L, n_pc=2, scale=TRUE,
   ME_diss_thres=0.25, qc_plot=TRUE)