lunaR

lunaR is a package for the R statistical software environment, which aims to provide a simple interface for the Luna library. Primary use cases are 1) for working with individual sleep studies/EDFs, and 2) as a convenient way to work with destrat output databases. For larger projects, the command-line version of Luna (lunaC) will likely be a better option.

Hint

To install lunaR, see the main installation pages, or try using the Dockerized version of Luna.

Overview

This page gives an overview of the core functions provided by this package, followed by examples and descriptions of some additional convenience functions. All examples on the page use the tutorial dataset, especially the second individual (nsrr02). The Luna tutorial also includes a section in which all steps are performed in lunaR as well as lunaC.

lunaR is simply an interface to the identical C/C++ library that underlies the lunaC command line tool: as such, lunaC and lunaR are fundamentally similar in most core respects. The different interfaces do mean that some aspects of the workflow differ between the two tools, however. For example, like lunaC, lunaR operates on a single EDF at a time; unlike the command line tool, however, lunaR is more interactive, in the sense that the modifiable, in-memory representation of the EDF typically persists between different lunaR commands. We discuss these differences below in more detail.

Note

In this documentation, we refer to Luna's R interface as lunaR to distinguish it from the base Luna library and command-line tool, which we're calling lunaC. The actual R package is named luna however (as a lunar package already exists in CRAN), as is the command line you'll actually type (i.e. luna) to run lunaC.

lunaR web-apps

We have built web-based applications on top of Luna, primarily:

Moonlight - an interactive browser tool for interactive viewing and analysis of a single recording
Moonbeam - an extension to Moonlight that enables NSRR users to directly pull NSRR data into Moonlight
Hypnoscope - a tool for viewing many hypnograms, with different options to sort, align and filter

Commands

The L-functions (i.e. all starting with the letter l for Luna) can be roughly grouped as follows:

attaching an EDF: lsl(), lattach(), ledf(), lstat(), ldrop(), lrefresh()
extracting data: lchs(), ldata(), ldata.intervals()
working with annotations: lannots(), lepoch(), letable(), lstages(), ladd.annot(), ladd.annot.file()
running commands: leval(), lcmd(), leval.project(), literate()
working with output: ldb(), lx(), lx2(), lid(), ltxttab()
setting variables: lset(), lvar(), lreset()
convenience/misc: lstgcols(), lbands(), llog(), le2i, lsanitize(), ldenoise(),
plotting functions: lheatmap(), ltopo.heat(), ltopo.rb() , ltopo.xy() , ltopo.heat2() , ltopo.topo() , ltopo.conn() , ltopo.dconn()

The index below gives a listing of all major lunaR functions:

Function	Description
`lsl()`	Loads a sample list
`lattach()`	Attaches an EDF from a sample list
`ledf()`	Attaches an EDF directly
`lstat()`	Reports on the currently attached EDF
`ldrop()`	Detaches the current EDF
`lrefresh()`	Reverts to the original attached EDF
`lchs()`	Returns a vector of channel names for the attached EDF
`ldata()`	Extracts signal (and annotation) data from an EDF for one or more epochs
`ldata.intervals()`	Extracts signal (and annotation) data from an EDF for one or more intervals
`lannots()`	Returns a vector of annotation class names for the attached EDF
`lepoch()`	Epochs the data
`letable()`	Returns an epoch-table with information about epochs, masks and annotations
`lstages()`	Return vector of sleep stages from annotations
`ladd.annot()`	Attaches an new annotation from an R interval list
`ladd.annot.file()`	Attaches an new annotation from an file
`leval()`	Evaluates an arbitrary set of Luna commands
`leval.project()`	Evaluates an arbitrary set of Luna commands for all members of a project
`lcmd()`	Reads and parses a Luna command script from a file
`literate()`	Applies an arbitrary R function to signal data, one epoch or interval at a time
`ldb()`	Imports data from a lunout database as an R list object
`lx()`	Extracts table(s) from objects returned by `ldb()`, `leval()` or `leval.project()
`lx2()`	Like `lx()` but for lists of returned objects
`lid()`	Extract a particular individual from a returned data frame
`ltxttab()`	Loads and conconcatenates multiple text-table output files
`lstgcols()`	Maps typical stage labels to colors (for plotting)
`lbands()`	Splits a signal into five band-pass filtered signals
`llog()`	Turns Luna's typical console messages on/off
`le2i()`	Converts epochs to intervals
`lsanitize()`	Cleans syntax of command, factor/level and channel names
`ldenoise()`	1D total variation denoiser
`lheatmap()`	Plots heatmaps, e.g. for spectrograms
`ltopo.heat()`	Topoplot (sensor-level)
`ltopo.rb()`	Topoplot (sensor-level) with red-white-blue
`ltopo.xy()`	Topoplot (sensor-level) X-Y plots
`ltopo.heat2()`	Topoplot (sensor-level) of X-Y-Z heatmaps
`ltopo.conn()`	Topoplot (sensor-level) of pairwise connectivity
`ltopo.dconn()`	Topoplot (sensor-level) of single-sensor seeded connectivity
`ltopo.topo()`	Topoplot (sensor-level) of topoplots
`ltopo.topo()`	Topoplot (sensor-level) of topoplots
`ltopo.topo()`	Topoplot (sensor-level) of topoplots
`ldefault.xy()`	Default XY channel locations for plotting
`ldefault.coh.xy()`	Default XY channel locations for coherence/connectivity plots