Draws important features heatmap in observations — draw_features

This shows the guide features scores in the observations, depending on whether or not they are in a given cluster. Each feature is scaled to a z-score.

draw_features_heatmap(
  stoic_results,
  data,
  cluster,
  obs = "cluster",
  only_positive = T
)

Arguments

data: Object returned by the `stoic_data` function.
cluster: Cluster of interest (numeric)
obs: value in "all", "cluster", "other". "all": shows all observations with a row annotation telling whether it is in the cluster "cluster": shows well predicted observations in the cluster only "other": shows all observations not in the cluster
only_positive: TRUE: Shows only the features that are important AND have higher values in the considered cluster. FALSE: shows all important features, including also the ones that are depleted in the cluster of interest as compared to other clusters.
stoic_clusters: Object returned by the `stoic_clustering` function.

Value

nothing, just shows the heatmap