Build a graph object from separate node and edge tables

Provides an explicit two-table interface where the user supplies a node table and an edge table separately. Compared with [build_graph_from_df()], which takes the edge list as the primary input and treats the node annotation as optional, here the node table is **required and authoritative**: nodes that have no incident edges are kept in the resulting graph (no orphan-node removal). This is the right entry point when the user already has a curated vertex set and wants every node to appear in the visualization, even if it ends up isolated after edge filtering.

Usage

build_graph_from_node_edge(
  node,
  edge,
  directed = FALSE,
  module.method = c("Fast_greedy", "Walktrap", "Edge_betweenness", "Spinglass"),
  top_modules = 15,
  seed = 1115
)

Arguments

node

Data frame. Node table whose **first column is the node identifier** that matches the `from` / `to` values in `edge`. Remaining columns are attached as vertex attributes (taxonomy, gene metadata, ...).

edge

Data frame. Edge list with at least two columns; the **first two columns are interpreted as `from` and `to`** (column names are not enforced). An optional `weight` column (any case) is used as edge weight; if no `weight`-named column is found and `ncol(edge) >= 3`, the third column is used.

directed

Logical (default: `FALSE`). Whether edges between nodes are directed.

module.method

Character. Network community detection method. Options: `"Fast_greedy"`, `"Walktrap"`, `"Edge_betweenness"`, `"Spinglass"`. `"Fast_greedy"` and `"Spinglass"` require an undirected graph.

top_modules

Integer. Number of top-ranked modules (by node count) to retain; smaller modules are collapsed into `"Others"`.

seed

Integer (default: `1115`). Random seed for reproducibility.

Value

A `tbl_graph` object with vertex columns `name`, `modularity`, `modularity2`, `modularity3`, `Modularity`, `Degree`, `Segree`, `Strength` plus any extra columns supplied via `node`, and edge columns `weight` (positive), `correlation` (signed) plus any extra columns supplied via `edge`.

Examples

if (FALSE) { # \dontrun{
node <- data.frame(
  name  = c("A", "B", "C", "D", "E"),
  group = c("g1", "g1", "g2", "g2", "g3")
)
edge <- data.frame(
  from   = c("A", "B", "C"),
  to     = c("B", "C", "D"),
  weight = c(0.8, 0.6, 0.7)
)
# Note: node "E" is isolated. Unlike build_graph_from_df(), the result
# of build_graph_from_node_edge() retains "E".
obj <- build_graph_from_node_edge(node = node, edge = edge)
obj
} # }