hypnoR — Hypnogram handling and sleep architecture metrics for R • hypnoR

Hypnogram handling, plotting, and sleep architecture metrics for R — the staging-agnostic layer of the Circadia Lab ecosystem.

⚠️ hypnoR is in early development and has not been formally tested. The API may change without notice, estimation results have not yet been validated against a reference implementation, and the package has not undergone peer review. Use with caution and verify outputs independently before using in any research context.

📖 What is hypnoR?

hypnoR is the hypnogram layer of the Circadia Lab R ecosystem. It sits between raw signal sources and the integrated participant database, providing a common interface for sleep staging regardless of where the staging came from.

It accepts two staging resolutions:

Full AASM (5-state: W / N1 / N2 / N3 / REM) — supplied by mrpheus
Coarse (3-state: W / Sleep / Quiet sleep) — supplied by zeitR

All metric and plotting functions are resolution-agnostic: they compute every metric that is possible given the available stages and return NA (documented per-function) for metrics that need finer staging than what’s available.

Downstream, hypnoR feeds into syncR::sync() as part of the unified participant-indexed database.

✨ Features

🏗️ new_hypnogram() — the constructor everything else goes through. Accepts a bare tibble with epoch/stage columns (e.g. zeitR::export_hypnogram() output) or an mrpheus_hypnogram object (mrpheus::export_hypnogram() output), auto-detecting AASM vs coarse resolution and normalising both into one representation.
🧹 smooth_hypnogram() — cleans up isolated single-epoch stage flips from raw, unsmoothed per-epoch classifiers, via two label-only rules (aasm_isolated, min_run), applicable independently or in sequence.
🪟 window_hypnogram() — restricts a hypnogram to a time (lights_off/lights_on) or epoch (from_epoch/to_epoch) window before any metric sees it.
📐 compute_sleep_architecture() — TST, TIB, SE, SOL, WASO, REM/SWS latency, and stage percentages; degrades gracefully for coarse hypnograms.
🌙 compute_cycles() — NREM/REM cycle segmentation via Feinberg & Floyd (1979) (method = "feinberg_floyd") or a gap-tolerant variant (method = "aasm"). Full AASM staging only.
🔀 compute_transitions() — stage-to-stage transition matrix (counts or probabilities) plus a fragmentation index and wake-transition count.
🎨 plot_hypnogram(), plot_architecture(), plot_transition_matrix() — publication-ready plots built on ggplot2 (a runtime-checked, not hard, dependency) and a shared Circadia Lab colour palette. plot_hypnogram() supports two visual styles: "step" (classic clinical trace) and "capsule" (rounded-pill bars per stage run, one lane per stage).

🚧 Coming soon

read_hypnogram() — direct ingestion from CSV, EDF annotations, YASA output, Compumedics Profusion, and Nox Medical files. Not yet implemented; for now, hypnograms come from an upstream package (mrpheus::export_hypnogram(), zeitR::export_hypnogram()) or a hand-built tibble passed to new_hypnogram().

🗂️ Project Structure

hypnoR/
├── R/
│   ├── hypnoR-package.R        # package-level docs
│   ├── new_hypnogram.R         # new_hypnogram() constructor
│   ├── smooth.R                # smooth_hypnogram()
│   ├── window_hypnogram.R      # window_hypnogram()
│   ├── architecture.R          # compute_sleep_architecture()
│   ├── cycles.R                # compute_cycles()
│   ├── transitions.R           # compute_transitions()
│   ├── read_hypnogram.R        # ingestion (not yet implemented)
│   ├── plot.R                  # plot_hypnogram(), plot_architecture(),
│   │                           #   plot_transition_matrix()
│   └── utils.R                 # internal helpers, stage colour palette
├── tests/testthat/
├── vignettes/
│   ├── getting-started.Rmd
│   └── articles/
│       ├── mrpheus-integration.Rmd  # worked example: AASM/mrpheus
│       └── zeitR-integration.Rmd    # worked example: coarse/zeitR
├── man/figures/                # logo, favicon, card
├── .github/workflows/          # R CMD CHECK + pkgdown/coverage CI
├── _pkgdown.yml
└── DESCRIPTION

🚀 Getting Started

Prerequisites: R ≥ 4.1.

Install from r-universe (recommended — pre-built binaries):

install.packages(
  "hypnoR",
  repos = c("https://circadia-bio.r-universe.dev", "https://cloud.r-project.org")
)

Or install the development version from GitHub:

# install.packages("remotes")
remotes::install_github("circadia-bio/hypnoR")

To reproduce the vignettes with real evaluated output (rather than just the code), also install mrpheus and zeitR from their r-universe:

install.packages(
  c("mrpheus", "zeitR"),
  repos = c("https://circadia-bio.r-universe.dev", "https://cloud.r-project.org")
)

Basic workflow (see vignette("getting-started") for the full walkthrough, including where staging/result come from in each case):

library(hypnoR)

# From mrpheus (full AASM)
hyp <- new_hypnogram(mrpheus::export_hypnogram(staging, epoch_s = 30))

# ...or from zeitR (coarse) -- same functions from here on, either way
hyp <- new_hypnogram(zeitR::export_hypnogram(result))

hyp   <- smooth_hypnogram(hyp)          # optional cleanup
arch  <- compute_sleep_architecture(hyp)
trans <- compute_transitions(hyp)
cyc   <- compute_cycles(hyp)            # full AASM only

plot_hypnogram(hyp, cycles = cyc)
plot_architecture(arch)
plot_transition_matrix(trans$matrix)

📦 Dependencies

Package	Type	Role
`cli`	Imports	User-facing messages and errors
`stats`	Imports	Faceting in `plot_architecture()`
`tibble`	Imports	Tidy output objects
`utils`	Imports	NSE variable declarations for `ggplot2`
`ggplot2`	Suggests	Plotting — checked at runtime, not required to use the metric functions
`mrpheus`, `zeitR`	Suggests	Real-data vignettes (both on circadia-bio’s r-universe)
`covr`, `testthat`	Suggests	Test coverage and the test suite
`knitr`, `rmarkdown`, `pkgdown`	Suggests	Vignettes and documentation site

👥 Authors

Role	Name
Author, maintainer	Lucas França
Author	Mario Leocadio-Miguel

Circadia Lab, Northumbria University.

📄 Citation

If you use hypnoR in your research, please cite it:

@software{franca_hypnor_2026,
  author  = {França, Lucas and Leocadio-Miguel, Mario},
  title   = {{hypnoR}: Hypnogram Handling, Plotting, and Sleep Architecture Metrics},
  year    = {2026},
  version = {0.1.1},
  doi     = {10.5281/zenodo.21309263},
  url     = {https://github.com/circadia-bio/hypnoR}
}

⌚️ zeitR — wrist actigraphy analysis and circadian metrics; upstream source of coarse hypnograms
🪉 mrpheus — PSG signal analysis; upstream source of full AASM hypnograms
🔄 syncR — integrates zeitR, slumbR, tallieR, and hypnoR into a unified participant-indexed database
🔬 circadia-bio — the Circadia Lab GitHub organisation

📄 Licence

Released under the MIT License.

😵‍💫 hypnoR