plotAbundance: Plotting abundance data
In microbiome/miaViz: Microbiome Analysis Plotting and Visualization

plotAbundance

R Documentation

Plotting abundance data

Description

plotAbundance plots the abundance on a selected taxonomic rank. Since this probably makes sense only for relative abundance data, the assay used by default is expected to be in the slot ‘relabundance’. If only ‘counts’ is present, the relative abundance is computed.

Usage

plotAbundance(x, ...)

## S4 method for signature 'SummarizedExperiment'
plotAbundance(
  x,
  rank = NULL,
  col.var = features,
  features = NULL,
  order.row.by = order_rank_by,
  order_rank_by = c("name", "abund", "revabund"),
  order.col.by = order_sample_by,
  order_sample_by = NULL,
  decreasing = TRUE,
  layout = c("bar", "point"),
  one.facet = one_facet,
  one_facet = TRUE,
  ncol = 2,
  scales = "fixed",
  assay.type = assay_name,
  assay_name = "counts",
  ...
)

Arguments

`x`	a `SummarizedExperiment` object.
`...`	additional parameters for plotting. `as.relative` `Character scalar`. Should the relative values be calculated? (Default: `FALSE`) See `mia-plot-args` for more details i.e. call `help("mia-plot-args")`
`rank`	`Character scalar`. Defines the taxonomic rank to use. Must be a value of `taxonomyRanks(x)`. (Default: `NULL`)
`col.var`	`Character scalar`. Selects a column from `colData` to be plotted below the abundance plot. Continuous numeric values will be plotted as point, whereas factors and character will be plotted as colour-code bar. (Default: `NULL`)
`features`	Deprecated. Use `col.var` instead.
`order.row.by`	`Character scalar`. How to order abundance value: By name (“name”) for sorting the taxonomic labels alphabetically, by abundance (“abund”) to sort by abundance values or by a reverse order of abundance values (“revabund”).
`order_rank_by`	Deprecated. Use `order.row.by` instead.
`order.col.by`	`Character scalar`. from the chosen rank of abundance data or from `colData` to select values to order the abundance plot by. (Default: `NULL`)
`order_sample_by`	Deprecated. Use `order.col.by` instead.
`decreasing`	`Logical scalar`. If the `order.col.by` is defined and the values are numeric, should the values used to order in decreasing or increasing fashion? (Default: `FALSE`)
`layout`	`Character scalar`. Either “bar” or “point”.
`one.facet`	`Logical scalar`. Should the plot be returned in on facet or split into different facet, one facet per different value detect in `rank`. If `col.var` or `order.col.by` is not `NULL`, this setting will be disregarded. (Default: `TRUE`)
`one_facet`	Deprecated. Use `one.facet` instead.
`ncol`	`Numeric scalar`. if `one.facet = FALSE`, `ncol` defines many columns should be for plotting the different facets. (Default: `2`)
`scales`	`Character scalar`. Defines the behavior of the scales of each facet. Both values are passed onto `facet_wrap`. (Default: `"fixed"`)
`assay.type`	`Character scalar` value defining which assay data to use. (Default: `"relabundance"`)
`assay_name`	Deprecate. Use `assay.type` instead.

Details

Subsetting to rows of interested and ordering of those is expected to be done outside of this functions, e.g. x[1:2,]. This will plot data of all col.var present.

Value

a ggplot object or list of two ggplot objects, if col.var are added to the plot.

Examples

data(GlobalPatterns, package="mia")
tse <- GlobalPatterns

## If rank is set to NULL (default), agglomeration is not done. However, note
## that there is maximum number of rows that can be plotted. That is why
## we take sample from the data.
set.seed(26348)
sample <- sample(rownames(tse), 20)
tse_sub <- tse[sample, ]
# Apply relative transformation
tse_sub <- transformAssay(tse_sub, method = "relabundance")
plotAbundance(tse_sub, assay.type = "relabundance")

## Plotting counts using the first taxonomic rank as default
plotAbundance(
    tse, assay.type="counts", rank = "Phylum") +
    labs(y="Counts")

## Using "Phylum" as rank. Apply relative transformation to "counts" assay.
plotAbundance(
    tse, assay.type="counts", rank = "Phylum", add_legend = FALSE,
    as.relative = TRUE)

# Apply relative transform
tse <- transformAssay(tse, method = "relabundance")
  
## A feature from colData or taxon from chosen rank can be used for ordering
## samples.
plotAbundance(tse, assay.type="relabundance", rank = "Phylum",
           order.col.by = "Bacteroidetes")

## col.var from colData can be plotted together with abundance plot.
# Returned object is a list that includes two plot; other visualizes
## abundance other col.var. 
plot <- plotAbundance(tse, assay.type = "relabundance", rank = "Phylum",
                   col.var = "SampleType")

# These two plots can be combined with wrap_plots function from patchwork
# package
library(patchwork)
wrap_plots(plot, ncol = 1)


## Same plot as above but showing sample IDs as labels for the x axis on the
## top plot
plot[[1]] <- plotAbundance(tse, assay.type = "relabundance", rank = "Phylum",
                           col.var = "SampleType", add.legend = FALSE,
                           add.x.text = TRUE)[[1]] +
                           theme(axis.text.x = element_text(angle = 90)) 

wrap_plots(plot, ncol = 1, heights = c(0.8,0.2))


## Compositional barplot with top 5 taxa and samples sorted by
## "Bacteroidetes"

# Getting top taxa on a Phylum level
tse <- transformAssay(tse, method="relabundance")
tse_phylum <- agglomerateByRank(tse, rank ="Phylum", onRankOnly=TRUE)
top_taxa <- getTop(tse_phylum,top = 5, assay.type = "relabundance")

# Renaming the "Phylum" rank to keep only top taxa and the rest to "Other"
phylum_renamed <- lapply(rowData(tse)$Phylum,
                       function(x){if (x %in% top_taxa) {x} else {"Other"}})
rowData(tse)$Phylum <- as.character(phylum_renamed)

# Compositional barplot
plotAbundance(tse, assay.type="relabundance", rank = "Phylum",
           order.row.by="abund", order.col.by = "Bacteroidetes")

microbiome/miaViz documentation built on Aug. 13, 2024, 8:57 p.m.