Plot the distribution of a numeric trait by group

Plots the values of a numeric trait as points, one row per group (e.g. dataset or family), with groups ordered by their mean value. The shape of each point shows whether it is a minimum, mean, maximum or raw value. Highlighted groups are drawn in pink and named in the bottom-right corner, so they can be found with hide_ids = TRUE. The x-axis is on a log scale when the values are all positive and the largest is more than about 13 times the smallest (so the axis, padded by 20% each side, spans more than a factor of 20). Values recorded as a bin or range are left out.

Usage

plot_trait_distribution_jitter(
  database,
  trait_name,
  y_axis_category,
  highlight = NA,
  hide_ids = FALSE
)

Arguments

database

traits.build database (list object), or its traits table (which must then have a family column, e.g. from join_taxa())

trait_name

Name of the trait to plot (matched exactly)

y_axis_category

Column to group values by, e.g. "dataset_id" or "family". family and genus are added from the taxa table when database is a whole database.

highlight

One or more groups to highlight, e.g. a dataset_id

hide_ids

Hide the group labels on the y-axis?

Value

A ggplot object, or a blank plot saying there are no numeric values to plot. Each point also carries a text aesthetic ("dataset_id | taxon_name"), which plotly::ggplotly() shows as a tooltip.

Author

Daniel Falster - daniel.falster@unsw.edu.au

Examples

austraits_lite %>% plot_trait_distribution_jitter("wood_density", "dataset_id", "Westoby_2014")