Skip to contents

Taxon trait summaries

Source: R/taxon_trait_summary.R
taxon_trait_summary.Rd

Allows evaluating summary functions (e.g. means, sd, median, mode, quantiles, etc.) on trait data by taxon

Usage

taxon_trait_summary(
  harmonized_trait_path,
  traits = NULL,
  taxonomic_level = "species",
  taxon_selection = NULL,
  summary_function = "weightedmean",
  summary_params = list(na.rm = TRUE),
  scalar_functions = NULL,
  priorization = TRUE,
  aggregation_level_weights = c(individual = 1, population = 10, taxon = 100),
  progress = TRUE,
  verbose = TRUE
)

Arguments

harmonized_trait_path

The path to harmonized trait data files (.rds or .csv format).

traits

A character vector with the set of traits to summarize. If NULL, then all available traits are summarized.

taxonomic_level

Taxonomic level for summary: either 'species' (acceptedName), 'genus' or 'family'

taxon_selection

String vector with taxon names to perform selection. By default, all taxa are returned.

summary_function

A function (statistic) to summarize numeric values for the same taxonomic entity (see details).

summary_params

A list of summary function params (by default na.rm=TRUE).

scalar_functions

A named list of scalar functions for traits needing transformation of units or scaling before summary.

priorization

A boolean flag to perform priorization of some data sources over others.

aggregation_level_weights

A vector of weights to be applied to different aggregation levels when calculating numeric averages.

progress

A boolean flag to prompt progress.

verbose

A boolean flag to prompt detailed process information.

Value

A data frame with taxon summary function values for the set of indicated traits.

Details

If priorization = TRUE and column priority_column is available in data sources, the function will prioritize sources with higher priority first, filling parameters with them before inspecting data sources of lower priority.

For categorical traits the following options are possible using summary_function and summary_params:

  • n - Number of non-missing observations.

  • weightedmode - Weighted mode using aggregation level weights.

  • mode - The usual statistic functions.

For numerical traits the following options are possible using summary_function and summary_params:

  • n - Number of non-missing observations.

  • weightedmean, weightedmedian, weightedquantile, weightedvar, weightedsd - Weighted statistics using aggregation level weights.

  • mean, median, quantile, var, sd - The usual statistic functions.

Examples

if (FALSE) { # \dontrun{
 # List harmonized trait files
 DB_path <- "~/OneDrive/mcaceres_work/model_development/medfate_parameterization/traits_and_models/"
 harmonized_trait_path <- paste0(DB_path,"data/harmonized_trait_sources")

 # Get family-level weighted means for two specific traits
 taxon_trait_summary(harmonized_trait_path, c("SLA", "Gswmax"),
                     taxonomic_level = "family", progress = FALSE)

} # }