TNA Plotting Gallery and Visual Smoke Test • cograph

This page is an executable visual smoke test for the complete plotting surface of tna and its integration with cograph. Knitting stops on the first plotting error. Statistical resampling uses deliberately small iteration counts so that the page can be rebuilt during development; the results below are for visual and API verification, not inference.

The gallery covers:

plot() for single and grouped TNA models, bootstrap objects, centralities, cliques, communities, comparisons, permutations, reliability, sequence comparisons, and stability objects;
hist(), plot_model(), plot_compare(), plot_frequencies(), plot_mosaic(), plot_sequences(), and plot_associations();
cograph’s enhanced TNA bootstrap, difference, permutation, forest, and heterogeneous-network renderers.

data.frame(
  package = c("tna", "cograph"),
  version = c(
    as.character(utils::packageVersion("tna")),
    as.character(utils::packageVersion("cograph"))
  )
)
#>   package version
#> 1     tna   1.2.3
#> 2 cograph   2.4.6

Shared fixtures

The fixtures use 400 rows from group_regulation, split into two groups. This is large enough to exercise real TNA objects while keeping resampling fast.

Core network plots

Single TNA model

Exercises plot.tna() and its default cograph renderer.

plot(model, title = "Single TNA model", minimum = 0.05)

Grouped TNA models

Exercises plot.group_tna().

plot(group_models, minimum = 0.05)

Heterogeneous TNA model

Passing a complete two-part node_list routes plot.tna() through cograph’s heterogeneous TNA renderer.

split_at <- ceiling(length(model$labels) / 2)
node_list <- list(
  model$labels[seq_len(split_at)],
  model$labels[seq.int(split_at + 1, length(model$labels))]
)
plot(
  model,
  node_list = node_list,
  title = "Heterogeneous TNA model",
  minimum = 0.05
)

Weight matrix

Exercises the exported plot_model() entry point.

tna::plot_model(
  model$weights,
  labels = model$labels,
  title = "TNA weight matrix"
)

Edge and state distributions

Edge-weight histograms

Exercises both hist.tna() and hist.group_tna().

hist(model, main = "Single-model edge weights")

hist(group_models)

State frequencies

Exercises plot_frequencies.tna() and plot_frequencies.group_tna().

tna::plot_frequencies(model)

tna::plot_frequencies(group_models)

Mosaic plots

Exercises mosaic methods for frequency TNA, grouped frequency TNA, and prepared tna_data objects.

tna::plot_mosaic(frequency_model)

tna::plot_mosaic(group_frequency_models)

tna::plot_mosaic(prepared_data, group = "cohort")

Association network

Exercises plot_associations.tna() using the required integer-valued frequency model.

tna::plot_associations(
  frequency_model,
  title = "Association network"
)

Sequence plots

Single-model index and distribution plots

Exercises plot_sequences.tna() in both modes.

tna::plot_sequences(model, type = "index", tick = 1)

tna::plot_sequences(model, type = "distribution", geom = "area", tick = 1)

Grouped sequences

Exercises plot_sequences.group_tna().

tna::plot_sequences(group_models, type = "distribution", tick = 1)

Data-frame sequences

Exercises plot_sequences.default() with an explicit grouping vector.

tna::plot_sequences(
  gallery_data,
  group = gallery_group,
  type = "distribution",
  tick = 1
)

Prepared TNA data

Exercises plot_sequences.tna_data() and metadata-based grouping.

tna::plot_sequences(
  prepared_data,
  group = "cohort",
  type = "index",
  tick = 1
)

Structural analysis plots

Centralities

Exercises plot.tna_centralities() and plot.group_tna_centralities().

plot(centrality_result, ncol = 3)

plot(group_centrality_result, ncol = 3)

Communities

Exercises plot.tna_communities() and plot.group_tna_communities().

plot(community_result, title = "Walktrap communities")

plot(group_community_result)

Cliques

Exercises plot.tna_cliques() and plot.group_tna_cliques(). Only the first dyad is drawn for each object to keep the gallery compact.

plot(clique_result, n = 1, ask = FALSE, title = "First dyad")

plot(group_clique_result, n = 1, ask = FALSE)

Bootstrap plots

TNA bootstrap method

Exercises plot.tna_bootstrap().

plot(bootstrap_result, title = "TNA bootstrap")

Grouped bootstrap method

Exercises plot.group_tna_bootstrap().

plot(group_bootstrap_result)

Cograph bootstrap render modes

These plots verify all enhanced splot.tna_bootstrap() display modes.

cograph::splot(
  bootstrap_result,
  display = "significant",
  title = "Bootstrap: significant edges",
  show_stars = TRUE
)

cograph::splot(
  bootstrap_result,
  display = "styled",
  title = "Bootstrap: styled full network",
  show_stars = TRUE
)

cograph::splot(
  bootstrap_result,
  display = "ci",
  title = "Bootstrap: confidence intervals",
  show_ci = TRUE
)

cograph::plot_bootstrap_forest(
  bootstrap_result,
  layout = "grouped",
  title = "Bootstrap forest grouped by source"
)

Permutation plots

TNA permutation method

Exercises plot.tna_permutation().

plot(permutation_result, title = "TNA permutation differences")

Grouped permutation method

Exercises plot.group_tna_permutation().

plot(group_permutation_result)

Cograph permutation renderer

Exercises the richer cograph permutation display directly.

cograph::plot_permutation(
  permutation_result,
  title = "Cograph permutation renderer",
  show_nonsig = TRUE
)

Stability and reliability plots

Centrality stability

Exercises plot.tna_stability().

plot(stability_result)

Grouped centrality stability

Exercises plot.group_tna_stability(). The method returns its ggplots invisibly, so the gallery prints each returned plot explicitly.

group_stability_plots <- plot(group_stability_result)
invisible(lapply(group_stability_plots, print))

Reliability distributions

Exercises all three modes of plot.tna_reliability().

plot(reliability_result, type = "histogram")

plot(reliability_result, type = "density")

plot(reliability_result, type = "boxplot")

Comparison plots

Difference networks

Exercises plot_compare.tna() and plot_compare.group_tna(), both of which delegate to cograph::plot_compare().

tna::plot_compare(
  model_first,
  model_second,
  title = "First half minus second half"
)

tna::plot_compare(group_models)

TNA comparison object

Exercises every mode of plot.tna_comparison().

plot(comparison_result, type = "heatmap")

plot(comparison_result, type = "scatterplot")

plot(comparison_result, type = "centrality_heatmap")

plot(comparison_result, type = "weight_density")

Sequence comparison

Exercises plot.tna_sequence_comparison().

plot(sequence_comparison_result, n = 10, cells = TRUE)

Build result

If this section is visible, every plotting chunk above completed without an error.

data.frame(
  status = "PASS",
  tna = as.character(utils::packageVersion("tna")),
  cograph = as.character(utils::packageVersion("cograph")),
  rendered_at = format(Sys.time(), tz = "UTC", usetz = TRUE)
)
#>   status   tna cograph             rendered_at
#> 1   PASS 1.2.3   2.4.6 2026-08-23 16:43:55 UTC