Generate a chart — make_krona • taxplore

This method generates the HTML code for a chart, which it returns or writes to a file.

Usage

make_krona(x, ...)

# S3 method for class 'taxonomyTable'
make_krona(x, tax_ranks = NULL, ...)

# S3 method for class 'phyloseq'
make_krona(
  x,
  magnitude = NULL,
  outfile = NULL,
  tax_ranks = NULL,
  dataset_group = NULL,
  group_vars = NULL,
  color = NULL,
  color_col = NULL,
  color_label = NULL,
  group_sep = " ",
  shorten_group = function(x) abbreviate(x, 60),
  ...
)

# S3 method for class 'data.frame'
make_krona(classification, ...)

# S3 method for class 'matrix'
make_krona(
  x,
  magnitude = NULL,
  outfile = NULL,
  dataset_group = NULL,
  color = NULL,
  color_label = NULL,
  color_value_range = NULL,
  attributes = NULL,
  dataset_summary_fn = sum,
  display = NULL,
  method = NULL,
  opts = NULL,
  ...
)

# S3 method for class 'list'
make_krona(
  x,
  magnitude = "n",
  outfile = NULL,
  datasets = NULL,
  attributes = NULL,
  color = NULL,
  color_value_range = NULL,
  opts = NULL,
  display = NULL,
  ...
)

Arguments

x

A classification matrix/data frame containing a hierarchical classification (e.g. taxonomic lineages), a phyloseq object or a tree data structure made of nested lists (see details below)

...

Passed on to make_krona.matrix, and remaining arguments are ultimately passed as ChartOpts as an alternative to opts for configuring HTML generation.

tax_ranks

Character vector with phyloseq taxonomy ranks names that should be chosen from the taxonomy table (default: all ranks)

magnitude

Specifies the magnitudes (abundances) of the different groups in the classification. If missing, equal magnitudes are assumed for all entries. Either a vector or matrix/data frame (see details)

outfile

Optional output file. If not provided, the HTML output is returned as character string.

dataset_group

Optional character vector specifying the grouping of magnitude columns or phyloseq samples into distinct datasets. By default, data in different columns of matrix-like abundance objects are summed up by row and globally displayed as one dataset. The dataset_group vector must be of length ncol(magnitude). The special value "separate" assumes that all all columns/phyloseq samples are separate data sets. Adjust the by-row summary function with dataset_summary_fn.

group_vars, group_sep, shorten_group

Character vector with phyloseq sample variables that define the grouping into different datasets (see dataset_group. Multiple variables are joined together using group_sep to form the dataset_group vector. Overly long strings are further shortened using the shorten_group function.

color_col

Name of a column in the phyloseq taxonomy table, which contains numeric values used for coloring (see color). Note that it is rather unusual to store extra (numeric) data in the taxonomy table, but it is still possible.

color_label

Label of the color scale (bottom left). Must be supplied if color is a data vector/matrix. Otherwise, the name or displayName of the ChartAttribute is used.

color_value_range

Lower and upper value limits that map to the color range (see hue_range in ChartOpts). By default, the full range() is used.

attributes

Optional list of ChartAttribute providing more information about nodes of in chart (when clicked), or for the coloring.

dataset_summary_fn

Function used to summarize the magnitudes from several samples (see dataset_group). sum is the default, but mean is also a good choice.

display

Optional ChartDisplayOpts configuring the initial appearance of the charts when embedded in RStudio or rendered documents. See set_chart_display_opts to apply these settings globally for multiple charts.

opts

Optional ChartOpts affecting how HTML charts are generated See also set_chart_opts to apply these settings globally for multiple charts.

datasets

If x is a tree data structure with magnitudes for >1 dataset (stored in 'n' attribute by default), then the names of these data sets need to be provided with datasets (character vector).

color:

Indicate data to be used for custom coloring of the chart. Should be one of:

Name of a ChartAttribute provided in the attributes list, which has numeric data to color by.
Numeric vector/matrix that directly provides the necessary data. See ChartAttribute for information about the data format and how values are aggregated within higher classification ranks. color_label also needs to be specified.

Value

Krona chart HTML code as character vector

Matrix-like classification

The classification (x) must be one of:

A matrix or data frame where each row defines a lineage of the hierarchy from left (higher ranks = inner circles in chart) to right (lower ranks = outer circles). Usually, an associated magnitudes vector or matrix/data frame is supplied along, indicating the size (weight) of the nodes in the chart. If not provided, equal weights are given to all nodes.
A Phyloseq object with at least a tax_table and otu_table component, or a simple taxonomyTable with or without additional magnitudes. Grouping into datasets can be done with group_vars (requires sample_data to be present).

magnitudes modify the width of the outermost nodes in the hierarchy (and the inner nodes along with them). It should be a vector of nrow(x) or a matrix/data frame matching the rows of x. By default, magnitude values from the different columns are summed up by row (configurable with dataset_summary_fn). Optionally samples can be grouped into datasets (see dataset_group). Any vector names/row names (if present) must match with nammes of x.

Phyloseq objects

If x is a Phyloseq object, no separate magnitude is required as abundances are taken from the otu_table() slot (but magnitude = ... can override it). Set magnitude = FALSE to give equal weight to all taxa.

Tree data structure

x can be a pre-assembled tree made of nested lists, which is written to the Krona XML unmodified. Each node requires at least a name attribute. The optional magnitude attribute (default name: magnitude = 'n') specifies the node weights for one or multiple datasets (specified with datasets). Additional data can be provided (see attributes).