splot()

The one-function network plot

Pass any network — matrix, igraph, tna, edge list, cograph — and get a publication-ready plot. Customize with 100+ parameters for nodes, edges, labels, donuts, pies, confidence intervals, themes, and more. All in base R graphics.

6 Input Formats 12+ Node Shapes Triple-Layer Donuts CI & P-value Overlays 7 Themes 100+ Parameters

Input Formats

100+

Parameters

12+

Node Shapes

Layout Algorithms

Built-in Themes

Donut + Pie Layers

Nodes

node_size · node_shape · node_fill · node_border_color · node_alpha · scale_nodes_by

18 parameters

Edges

edge_color · edge_width · edge_alpha · curvature · arrow_size · edge_style

20 parameters

Donuts & Pies

donut_fill · donut_color · pie_values · pie_colors · donut2_values · donut_shape

22 parameters

Edge Labels

edge_labels · edge_label_style · edge_label_template · edge_label_p · edge_label_stars

25 parameters

Layout & Theme

layout · theme · background · seed · layout_scale · rescale

14 parameters

Output

filetype · filename · width · height · res · title

7 parameters

# That's it. One line.
splot(model)

Input & Layout

Accepted Formats

splot() auto-detects the input type. Directionality is inferred from matrix symmetry or the source object. Override with directed.

Parameter	Default	Description
`x`	required	Network data: matrix, data.frame (edge list), igraph, cograph_network, tna, group_tna, bootstrap, or permutation object
`directed`	`NULL`	Force directed/undirected. `NULL` = auto-detect from input
`i`	`NULL`	Group index or name when `x` is group_tna. `NULL` = plot all groups in a grid

Tip: Matrices with symmetric weights are treated as undirected. Edge lists default to directed. igraph/tna objects carry their own directionality.

Layout Algorithms

Parameter	Default	Description
`layout`	`"oval"`	Layout algorithm or coordinate matrix
`seed`	`42`	Random seed for reproducible layouts
`rescale`	`TRUE`	Rescale layout to −1 to 1 range
`aspect`	`TRUE`	Maintain aspect ratio

Built-in layouts: "oval", "circle", "spring", "groups", "grid", "star", "bipartite". Or pass any igraph layout by name ("fr", "kk", "drl", "mds", "ni", "tr") or as a two-column matrix of x,y coordinates.

Example: Layouts

splot(mat, layout = "circle")
splot(mat, layout = "spring", seed = 123)
splot(mat, layout = "groups", groups = c("A","A","B","B","B"))

# igraph layout
splot(mat, layout = "kk")

# Custom coordinates
coords <- cbind(x = runif(10), y = runif(10))
splot(mat, layout = coords)

Plot Controls

Parameter	Default	Description
`title`	`NULL`	Plot title
`title_size`	`1.2`	Title font size
`background`	`"white"`	Background color
`layout_scale`	`1`	Scale factor: >1 expands, <1 contracts. `"auto"` scales by node count
`layout_margin`	`0.15`	Margin as fraction of layout range
`margins`	`c(.1,.1,.1,.1)`	Plot margins: c(bottom, left, top, right)
`scaling`	`"default"`	`"default"` (qgraph-matched) or `"legacy"` (pre-v2.0)

Node Aesthetics

Size, Shape & Color

Parameter	Default	Description
`node_size`	`NULL`	Node size. Scalar or per-node vector. Default 3
`node_size2`	`NULL`	Secondary size for ellipse/rectangle height
`node_shape`	`"circle"`	Shape name or per-node vector
`node_svg`	`NULL`	Custom SVG: file path or inline string
`svg_preserve_aspect`	`TRUE`	Maintain SVG aspect ratio
`node_fill`	`NULL`	Fill color(s). Uses theme default if NULL
`node_border_color`	`NULL`	Border color(s)
`node_border_width`	`1`	Border width(s)
`node_alpha`	`1`	Transparency (0–1)
`use_pch`	`FALSE`	Use `points()` for simple circles (faster for large networks)

Available shapes: "circle", "square", "triangle", "diamond", "pentagon", "hexagon", "star", "heart", "ellipse", "cross", "rectangle", "pie", plus any registered SVG shape.

Example: Node styling

# Per-node shapes and colors
splot(mat,
  node_shape = c("circle", "hexagon", "diamond", "square", "star"),
  node_fill = c("#E63946", "#457B9D", "#2A9D8F", "#E9C46A", "#F4A261"),
  node_size = 8,
  node_border_width = 2
)

# Custom SVG nodes
splot(mat, node_svg = "path/to/icon.svg", node_size = 10)

Centrality Scaling

Size nodes proportionally to any centrality measure. The value is computed internally — just pass the measure name.

Parameter	Default	Description
`scale_nodes_by`	`NULL`	Centrality measure name: `"degree"`, `"strength"`, `"betweenness"`, `"closeness"`, `"eigenvector"`, `"pagerank"`, `"authority"`, `"hub"`, `"harmonic"`, `"indegree"`, `"outdegree"`, `"instrength"`, `"outstrength"`, …
`node_size_range`	`c(2, 8)`	Output size range c(min, max)
`scale_nodes_scale`	`1`	Dampening exponent. <1 compresses differences, >1 exaggerates

Example: Size by centrality

# PageRank scaling
splot(mat, scale_nodes_by = "pagerank", node_size_range = c(3, 12))

# In-strength with dampening
splot(mat, scale_nodes_by = "instrength", scale_nodes_scale = 0.5)

Labels

Parameter	Default	Description
`labels`	`TRUE`	`TRUE` (use names), `FALSE` (none), or character vector
`label_size`	`NULL`	Label text size (cex)
`label_color`	`"black"`	Label color
`label_position`	`"center"`	`"center"`, `"above"`, `"below"`, `"left"`, `"right"`
`label_fontface`	`"plain"`	`"plain"`, `"bold"`, `"italic"`, `"bold.italic"`
`label_fontfamily`	`"sans"`	`"sans"`, `"serif"`, `"mono"`
`label_hjust`	`0.5`	Horizontal justification (0=left, 1=right)
`label_vjust`	`0.5`	Vertical justification (0=bottom, 1=top)
`label_angle`	`0`	Rotation angle in degrees

Donuts & Pies

Nodes can carry up to three data layers: an outer donut ring, an inner donut ring, and a pie chart core. Use these to encode multiple variables per node without cluttering the graph.

Donut Rings

Parameter	Default	Description
`donut_fill`	`NULL`	Fill proportion (0–1) per node. Scalar or vector
`donut_color`	`NULL`	Fill color(s) for the ring. Scalar, vector, or list of vectors for segmented donuts
`donut_inner_ratio`	`0.8`	Inner radius as fraction of outer (0–1). Smaller = thicker ring
`donut_bg_color`	`"gray90"`	Background color for unfilled portion
`donut_shape`	`"circle"`	Base shape: `"circle"`, `"square"`, `"hexagon"`, `"triangle"`, `"diamond"`, `"pentagon"`
`donut_border_color`	`NULL`	Border color. NULL inherits from `node_border_color`
`donut_border_width`	`NULL`	Border width. NULL inherits from `node_border_width`
`donut_outer_border_color`	`NULL`	Outer boundary color (enables double border)
`donut_line_type`	`"solid"`	Line type: `"solid"`, `"dashed"`, `"dotted"`
`donut_empty`	`TRUE`	Render empty rings for NA values

Donut Value Display

Parameter	Default	Description
`donut_show_value`	`FALSE`	Show numeric value in donut center
`donut_value_size`	`0.8`	Font size for center value
`donut_value_color`	`"black"`	Color for center value
`donut_value_fontface`	`"bold"`	Font face for center value
`donut_value_digits`	`2`	Decimal places
`donut_value_prefix`	`""`	Text before value (e.g., `"$"`)
`donut_value_suffix`	`""`	Text after value (e.g., `"%"`)

Example: Donut rings

# Simple progress ring
splot(mat, donut_fill = c(0.9, 0.7, 0.5), donut_color = "steelblue")

# Segmented donut (3 categories per node)
splot(mat,
  donut_fill = lapply(1:5, function(i) c(runif(1), runif(1), runif(1))),
  donut_color = list(c("#E63946", "#457B9D", "#2A9D8F")),
  donut_inner_ratio = 0.6
)

# Hexagonal donut with value display
splot(mat,
  donut_fill = c(0.85, 0.72, 0.93),
  donut_color = "coral",
  donut_shape = "hexagon",
  donut_show_value = TRUE,
  donut_value_suffix = "%",
  donut_value_digits = 0
)

Pie Charts

Parameter	Default	Description
`pie_values`	`NULL`	List of numeric vectors: segment sizes per node
`pie_colors`	`NULL`	Color vector or list of vectors for pie segments
`pie_border_width`	`NULL`	Border width for slice dividers

Example: Pie + Donut combo

splot(mat,
  node_shape = "donut",
  donut_fill = c(0.9, 0.7, 0.5),
  donut_color = "steelblue",
  pie_values = list(c(3,2,1), c(1,1,4), c(2,3,1)),
  pie_colors = c("#E41A1C", "#377EB8", "#4DAF4A"),
  node_size = 8
)

Double Donut

A second donut ring inside the first, for encoding a third data dimension.

Parameter	Default	Description
`donut2_values`	`NULL`	Values for inner donut ring
`donut2_colors`	`NULL`	Colors for inner donut ring
`donut2_inner_ratio`	`0.4`	Inner radius ratio for inner donut

Example: Triple layer (donut + donut + pie)

splot(mat,
  donut_fill = lapply(1:5, function(i) runif(3)),
  donut_color = list(c("#E63946", "#457B9D", "#2A9D8F")),
  donut_inner_ratio = 0.7,
  donut2_values = as.list(runif(5, 0.4, 0.9)),
  donut2_colors = list("#F4A261"),
  donut2_inner_ratio = 0.45,
  pie_values = lapply(1:5, function(i) sample(1:5, 4)),
  pie_colors = list(c("#264653", "#E9C46A", "#F4A261", "#E76F51")),
  node_size = 7
)

Edge Aesthetics

Color, Width & Style

Parameter	Default	Description
`edge_color`	`NULL`	Edge color(s). NULL = positive/negative coloring by weight sign
`edge_positive_color`	`"#2E7D32"`	Color for positive weights
`edge_negative_color`	`"#C62828"`	Color for negative weights
`edge_width`	`NULL`	Edge width(s). NULL = scale by weight
`edge_alpha`	`0.8`	Transparency (0–1)
`edge_style`	`1`	Line type: 1=solid, 2=dashed, 3=dotted, etc.

Weight Scaling

How edge weights map to visual width. The default is adaptive — it adjusts to network size and weight distribution.

Parameter	Default	Description
`edge_size`	`NULL`	Base size for weight scaling. NULL = adaptive: `15 * exp(-n/90) + 1`
`edge_width_range`	`c(0.1, 4)`	Output width range c(min, max)
`edge_scale_mode`	`"linear"`	Scaling: `"linear"`, `"log"`, `"sqrt"`, or `"rank"`
`edge_cutoff`	`NULL`	Two-tier width threshold. NULL = auto (75th percentile), 0 = disabled

Note: edge_scale_mode = "log" is useful when weights span several orders of magnitude. "rank" equalizes spacing and is useful for heavy-tailed distributions.

Arrows & Curvature

Parameter	Default	Description
`curvature`	`0`	Edge curvature. 0 = straight
`curve_scale`	`TRUE`	Auto-curve reciprocal edges
`curve_shape`	`0`	Spline tension (−1 to 1)
`curve_pivot`	`0.5`	Control point position along edge (0–1)
`curves`	`TRUE`	`TRUE` = auto-curve reciprocals, `FALSE` = all straight, `"force"` = all curved
`arrow_size`	`1`	Arrow head size
`arrow_angle`	`pi/6`	Arrow head angle (radians). Default = 30°
`show_arrows`	`TRUE`	Show arrows on directed edges. Logical or per-edge vector
`bidirectional`	`FALSE`	Arrows at both ends. Logical or per-edge vector
`loop_rotation`	`NULL`	Self-loop direction angle(s) in radians

Edge Start Style

A visual cue at the source end of directed edges — a dashed or dotted segment that indicates direction without relying solely on arrowheads.

Parameter	Default	Description
`edge_start_style`	`"solid"`	`"solid"`, `"dashed"`, or `"dotted"`
`edge_start_length`	`0.15`	Fraction of edge length for styled segment (0–0.5)
`edge_start_dot_density`	`"12"`	Dot pattern: two-char string (dot length, gap length)

Edge Labels

Basic Labels

Parameter	Default	Description
`edge_labels`	`FALSE`	`TRUE` (show weights), `FALSE` (none), or character vector
`edge_label_size`	`0.8`	Label text size
`edge_label_color`	`"gray30"`	Text color
`edge_label_bg`	`NA`	Background color for label box
`edge_label_position`	`0.5`	Position along edge (0=source, 1=target)
`edge_label_offset`	`0`	Perpendicular offset (positive = above curve)
`edge_label_fontface`	`"plain"`	Font face
`edge_label_halo`	`TRUE`	White halo for readability
`edge_label_digits`	`2`	Decimal places for numeric values
`edge_label_leading_zero`	`TRUE`	Show leading zero for values < 1
`edge_label_oneline`	`TRUE`	Single line format

Shadow

Parameter	Default	Description
`edge_label_shadow`	`FALSE`	Enable drop shadow
`edge_label_shadow_color`	`"gray40"`	Shadow color
`edge_label_shadow_offset`	`0.5`	Shadow offset in points
`edge_label_shadow_alpha`	`0.5`	Shadow transparency

Templates & Styles

The template system composes edge labels from multiple data sources. Use preset styles or build your own with placeholders.

Parameter	Default	Description
`edge_label_style`	`"none"`	Preset: `"none"`, `"estimate"`, `"full"`, `"range"`, `"stars"`
`edge_label_template`	`NULL`	Custom template with placeholders. Overrides `edge_label_style`

Template placeholders: {est} = estimate, {low} = lower CI, {up} = upper CI, {range} = [low, up], {p} = p-value, {stars} = significance stars.

Style	Produces
`"estimate"`	`.23`
`"full"`	`.23 [.18, .29]*`
`"range"`	`[.18, .29]`
`"stars"`	`***`

Example: Custom template

# Effect size with stars
splot(mat,
  edge_labels = TRUE,
  edge_label_template = "{est}{stars}",
  edge_label_p = pval_matrix,
  edge_label_stars = TRUE,
  edge_label_leading_zero = FALSE
)

# Full annotation
splot(mat,
  edge_label_style = "full",
  edge_ci_lower = ci_lo,
  edge_ci_upper = ci_hi,
  edge_label_p = pvals,
  edge_label_stars = TRUE
)

Confidence Intervals & P-values

Overlay confidence intervals as semi-transparent bands on edges, and annotate with p-values and significance stars.

Parameter	Default	Description
`edge_ci`	`NULL`	CI width vector (0–1). Rendered as semi-transparent underlay
`edge_ci_scale`	`2.0`	Width multiplier for underlay thickness
`edge_ci_alpha`	`0.15`	Underlay transparency
`edge_ci_color`	`NA`	Underlay color. NA = match edge color
`edge_ci_style`	`2`	Underlay line type: 1=solid, 2=dashed, 3=dotted
`edge_ci_arrows`	`FALSE`	Show arrows on underlay
`edge_ci_lower`	`NULL`	Lower CI bounds (for label annotation)
`edge_ci_upper`	`NULL`	Upper CI bounds (for label annotation)
`edge_label_p`	`NULL`	P-values per edge
`edge_label_p_digits`	`3`	Decimal places for p-values
`edge_label_p_prefix`	`"p="`	Prefix for p-value display
`edge_label_stars`	`NULL`	Significance stars: `TRUE` (compute from p), character vector, or numeric (treated as p-values)
`edge_label_ci_format`	`"bracket"`	`"bracket"` [low, up] or `"dash"` low–up
`edge_priority`	`NULL`	Render priority per edge. Higher = drawn on top

Example: Full statistical annotation

splot(mat,
  edge_ci_lower = ci_lower_matrix,
  edge_ci_upper = ci_upper_matrix,
  edge_label_p = pvalue_matrix,
  edge_label_stars = TRUE,
  edge_label_style = "full",
  edge_label_leading_zero = FALSE,
  edge_label_color = "#800020"
)

Weight Handling

Parameter	Default	Description
`threshold`	`0`	Minimum absolute weight to display an edge
`minimum`	`0`	Alias for `threshold` (qgraph compatibility). Uses `max(threshold, minimum)`
`maximum`	`NULL`	Maximum weight for scaling. NULL = auto from data
`weight_digits`	`2`	Decimal places to round weights. Edges rounding to 0 are removed. NULL = disable
`edge_duplicates`	`NULL`	Handle duplicate edges in undirected networks: `"sum"`, `"mean"`, `"first"`, `"max"`, `"min"`, or custom function

Themes & Legend

Built-in Themes

Parameter	Default	Description
`theme`	`NULL`	Theme name. NULL = default styling

Available themes: "classic" (white bg, blue nodes), "dark" (dark bg for slides), "minimal" (subtle, thin edges), "colorblind" (CVD-optimized), "gray" (grayscale), "nature" (earth tones), "viridis" (viridis palette). Register custom themes with register_theme().

Legend

Parameter	Default	Description
`legend`	`FALSE`	Show legend
`legend_position`	`"topright"`	`"topright"`, `"topleft"`, `"bottomright"`, `"bottomleft"`
`legend_size`	`0.8`	Legend text size
`legend_edge_colors`	`TRUE`	Show positive/negative edge colors
`legend_node_sizes`	`FALSE`	Show node size scale
`groups`	`NULL`	Group assignments for node coloring and legend entries
`node_names`	`NULL`	Alternative names for legend (separate from plot labels)

Output & Saving

Parameter	Default	Description
`filetype`	`"default"`	`"default"` (screen), `"png"`, `"pdf"`, `"svg"`, `"jpeg"`, `"tiff"`
`filename`	`tempdir()`	Output path (without extension)
`width`	`7`	Width in inches
`height`	`7`	Height in inches
`res`	`600`	Resolution (DPI) for raster formats

Example: Save to file

# Save as PDF
splot(mat, filetype = "pdf", filename = "network", width = 10, height = 8)

# Save as high-res PNG
splot(mat, filetype = "png", filename = "network", width = 10, height = 8, res = 300)

S3 Methods: splot has specialized methods for tna_bootstrap (solid/dashed by significance, stars on labels), tna_permutation (green/red by group direction), and group_tna_permutation objects. These add parameters like display, show_ci, show_stars, and width_by on top of the full parameter set above.