build_net: Network building by co-expression score computation

In GWENA: Pipeline for augmented co-expression analysis

Description

Compute the adjacency matrix, then the TOM to build the network. Than detect the modules by hierarchical clustering and thresholding

Usage

build_net(
  data_expr,
  fit_cut_off = 0.9,
  cor_func = c("pearson", "spearman", "bicor", "other"),
  your_func = NULL,
  power_value = NULL,
  block_size = NULL,
  stop_if_no_fit = FALSE,
  network_type = c("unsigned", "signed", "signed hybrid"),
  tom_type = c("unsigned", "signed", "signed Nowick", "unsigned 2", "signed 2", "none"),
  keep_matrices = c("none", "cor", "adja", "both"),
  n_threads = NULL,
  ...
)

Arguments

data_expr	matrix or data.frame or SummarizedExperiment, expression data with genes as column and samples as row.
fit_cut_off	float, cut off by which R^2 (coefficient of determination) will be thresholded. Must be in ]0;1[.
cor_func	string, name of the correlation function to be used. Must be one of "pearson", "spearman", "bicor", "other". If "other", your_func must be provided
your_func	function returning correlation values. Final values must be in [-1;1]
power_value	integer, power to be applied to the adjacency matrix. If NULL, will be estimated by trying different power law fitting.
block_size	integer, size of blocks by which operations can be proceed. Helping if working with low capacity computers. If null, will be estimated.
stop_if_no_fit	boolean, does not finding a fit above fit_cut_off should stop process, or just print a warning and return the highest fitting power.
network_type	string, type of network to be used. Either "unsigned", "signed", "signed hybrid". See details.
tom_type	string, type of the topological overlap matrix to be computed. Either "none", "unsigned", "signed", "signed Nowick", "unsigned 2", "signed 2" and "signed Nowick 2". See detail at TOMsimilarityFromExpr.
keep_matrices	string, matrices to keep in final object. Can be one of "none", "cor", "adja", "both". It is usefull to keep both if you plant to use compare_conditions.
n_threads	integer, number of threads that can be used to paralellise the computing
...	any other parameter compatible with adjacency.fromSimilarity

Value

list containing network matrix, metadata of input parameters and power fitting information.

Examples

net <- build_net(kuehne_expr[, seq_len(350)], n_threads = 1)

GWENA documentation built on Feb. 17, 2021, 2:01 a.m.