PlotClustersOverlay: Create a clusters-overlay plot

In davnovak/SingleBench: Framework for benchmarking and optimising dimension-reduction and clustering pipelines in expression data analysis

Description

Creates a plot showing separation of generated clusters using previously generated 2-dimensional layout of input expression data. For large numbers of data points, this function uses raster graphics to visualise the layout faster.

Usage

PlotClustersOverlay(
  benchmark,
  idx.subpipeline,
  idcs.n_param = NULL,
  exclude_unassigned = FALSE,
  raster_threshold = 5000L,
  plot_title = "Clustering"
)

Arguments

benchmark	object of type Benchmark
idx.subpipeline	integer: index of sub-pipeline that includes a clustering step
idcs.n_param	integer or vector of integers: indices of n-parameter (cluster count) of interest. Default value is NULL, which translates to all n-parameter iterations of the given sub-pipeline
exclude_unassigned	logical: if TRUE, data points that are considered unassigned per manual annotation are omitted. Default value is FALSE
raster_threshold	integer: maximum number of data points for which vector graphics should be used. Default value is 5000
plot_title	string: title of the plot. Default value is 'Clustering'

Details

You need to specify a single sub-pipeline and one or more n-parameter iteration by index. If multiple n-parameter iterations are chosen, a list of plots will be generated.

You can choose a custom title for your plot. The plot subtitle is always the n-parameter iteration name.

See Also

• Benchmark: constructs a Benchmark object (benchmark pipeline set-up)\

• Evaluate: runs all benchmark sub-pipelines and scores the performance of each tool

• AddLayout: allows you to add a separate 2-dimensional layout of the input dataset or to use an existing projection (produced in the evaluation) as a visualisation layout.

davnovak/SingleBench documentation built on Dec. 19, 2021, 9:10 p.m.