---
title: "Extended Sequence Visualisations"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Extended Sequence Visualisations}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(collapse = TRUE, comment = "#>")
library(gp3sequences)
```

## Synthetic data

```{r data}
data <- data.frame(
  sequence_id = rep(paste0("s", 1:8), each = 6L),
  sequence_order = rep(1:6, times = 8L),
  state = c(
    rep(c("A", "B", "B", "C", "D", "D"), 4L),
    rep(c("D", "C", "C", "B", "A", "A"), 4L)
  ),
  stringsAsFactors = FALSE
)
distance <- compute_sequence_distance(data, method = "levenshtein")
clustering <- cluster_sequences(distance, k = 2L, method = "hierarchical")
network <- create_transition_network(data)
```

## Sequence index

```{r index, fig.width=7, fig.height=5}
plot_sequence_index(data)
```

## State distribution and entropy

```{r distribution, fig.width=7, fig.height=4}
plot_sequence_state_distribution(data)
```

```{r entropy, fig.width=7, fig.height=4}
plot_sequence_entropy(data)
```

Entropy is a structural diversity summary at each aligned position. It is not a
measure of participant uncertainty or cognition.

## Distance and clustering diagnostics

```{r distance, fig.width=6, fig.height=5}
plot_sequence_distance_heatmap(distance)
```

```{r silhouette, fig.width=7, fig.height=4}
plot_sequence_cluster_silhouette(clustering, distance)
```

## Transition network

```{r network, fig.width=6, fig.height=6}
plot_transition_network(network)
```

These base-R plots are intentionally focused on package-native audited objects.
They complement, rather than replace, specialist visualisation ecosystems.