Skip to contents

Create sample-level subgraph topology in parallel

Source: R/get_sample_subgraph_topology_parallel.R
get_sample_subgraph_topology_parallel.Rd

For each sample (each column in mat), OTUs with non-zero abundance are treated as present (OTU != 0). The function extracts a subgraph from graph_obj using igraph::subgraph(), then computes topology for each sample-specific subgraph via get_network_topology().

Usage

get_sample_subgraph_topology_parallel(
  graph_obj,
  mat = NULL,
  transfrom.method = c("none", "scale", "center", "log2", "log10", "ln", "rrarefy",
    "rrarefy_relative"),
  r.threshold = 0.7,
  p.threshold = 0.05,
  method = c("WGCNA", "SpiecEasi", "SPARCC", "cor"),
  cor.method = c("pearson", "kendall", "spearman"),
  proc = c("holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none"),
  SpiecEasi.method = c("mb", "glasso"),
  sparcc_R = 20,
  bootstrap = 100,
  parallel = FALSE,
  n_workers = NULL,
  seed = 1115
)

Arguments

graph_obj

An graph object from build_graph_from_mat or build_graph_from_df.

mat

Numeric matrix used to build graph_obj. Rows are OTUs and columns are samples.

transfrom.method

Character. Passed to get_network_topology().

r.threshold

Numeric. Passed to get_network_topology().

p.threshold

Numeric. Passed to get_network_topology().

method

Character. Passed to get_network_topology().

cor.method

Character. Passed to get_network_topology().

proc

Character. Passed to get_network_topology().

sparcc_R

Integer. Passed to get_network_topology() for SparCC p-values. Default 20.

bootstrap

Numeric (default = 100). Passed to get_network_topology().

parallel

Logical (default = FALSE). Whether to enable parallel computation across samples.

n_workers

Integer (default = NULL). Number of workers when parallel = TRUE. If NULL, use future::availableCores() - 1.

seed

Integer (default = 1115). Random seed for reproducibility.

Value

A list with:

  • subgraph_list: sample-wise tbl_graph subgraphs

  • topology: merged sample-wise topology table

  • Robustness: merged sample-wise robustness table

  • sample_stat: sample-wise node/edge counts and status

Examples

NULL
#> NULL