Once the scores of interest have been computed and saved in the score tibble, they can be scaled to make them comparable by bounding them between 0 and 1 and harmonise their direction (0 and 1 always mean bad and good performance respectively). This is also a prerequisite for plotting.
Usage
ScaleScores(
object,
ref = "Unintegrated",
rescale = c("none", "rank", "score"),
batch.coeff = 0.4,
bio.coeff = 0.6
)Arguments
- object
a Seurat object
- ref
the name of the integration to use as a reference for scaling. Useful for PCA regression (and density) and cell cycle conservation scores.
- rescale
whether to rescale each score between 0 and 1 using min-max normalisation before computing overall scores. One of 'none' (default, no rescaling), 'rank' or 'score', corresponding to no min-max normalisation of ranks or scores respectively. Beware that the argument of
PlotScores()overrides the choice here (see Details sections).- batch.coeff
the weight of batch correction performance evaluation scores in the overall score.
- bio.coeff
the weight of bio-conservation performance evaluation scores in the overall score.
Details
It is recommended to keep the default parameter rescale = "none" to
retain the possibility to plot scores without min-max rescaling later with
PlotScores().
See also
PlotScores() for further details on rescaling