Convert a PhysioExperiment to an MNE-Python RawArray

Builds an mne.io.RawArray from a PhysioExperiment: channel names, channel types and sampling rate populate the mne.Info; any electrode positions (getElectrodePositions()) become a DigMontage; and any events (getEvents()) become mne.Annotations. The signal matrix is passed through unchanged, so fromMNE(toMNE(x)) reproduces it exactly.

Usage

toMNE(x, assay = NULL)

Arguments

x

A PhysioExperiment.

assay

Assay to export (default: the object's default assay).

Value

An mne.io.RawArray Python object.

Details

Channel types not recognised by MNE are mapped to "misc" (and are lower-cased to MNE's convention). Channel names must be unique. An event's type and (when present) value are packed into the annotation description so both survive the round-trip; a missing (NA) value becomes "" because MNE annotations cannot represent NA. Events whose onset falls outside the recording are dropped by MNE (with a warning).

References

Gramfort A, et al. (2013). "MEG and EEG data analysis with MNE-Python." Frontiers in Neuroscience, 7, 267.

See also

fromMNE(), hasMNE()

Examples

if (hasMNE()) {
  pe <- PhysioExperiment(
    assays = S4Vectors::SimpleList(raw = matrix(rnorm(300), 100, 3)),
    colData = S4Vectors::DataFrame(label = c("Fp1", "Fp2", "Cz"),
                                   type = rep("eeg", 3)),
    samplingRate = 100)
  raw <- toMNE(pe)
}