EcoDiet (version 2.0.1)

preprocess_data: Check and preprocess the data

Description

This function preprocesses the data input by the user, checks that the different inputs have the right format, and creates the data list that will feed the JAGS model.

If an error appears with a clear message, it means that the input needs to be reformatted. Please follow the instructions in the error message. You can also look at the data examples to guide you.

Usage

preprocess_data(
  biotracer_data,
  trophic_discrimination_factor,
  literature_configuration = FALSE,
  topology = NULL,
  element_concentration = 1,
  stomach_data = NULL,
  rescale_stomach = FALSE,
  literature_diets = NULL,
  nb_literature,
  literature_slope
)

Value

A list of preprocessed data, ready to be fed to the EcoDiet model through the 'rjags' package. This list contains a different number of elements, depending on which configuration is chosen by the user (literature_configuration = TRUE or FALSE).

Arguments

biotracer_data

A dataframe containing the biotracer data in the specific format: the first column corresponds to the trophic group or latin species and the remaining columns contains the biotracer measures

trophic_discrimination_factor

A vector containing the trophic discrimination factors corresponding to each column found in the biotracer data (except the group column of course)

literature_configuration

A boolean (TRUE or FALSE) indicating whether the model will have prior distributions informed by a literature study

topology

A matrix that the user may input if she wants the model to investigate some additionnal trophic links (by-default it is set on NULL and defined later from the stomach content data and the literature diets if present)

element_concentration

A matrix containing the element concentration for each trophic group and each biotracer element (listed in the biotracer data). It is a matrix with as many columns as the number of trophic groups and as many rows as the number of elements. By default the matrix is filled with ones.

stomach_data

A dataframe containing the stomach content data in a specific format: the first row contains the names of the prey trophic groups, the headers contains the names of the consumer / predator trophic groups, and the rest are the number of the predator's stomachs in which this prey was found. The last row contains the total number of non-empty stomach for the corresponding predator.

rescale_stomach

A boolean (TRUE or FALSE) indicating whether the stomach content data will be rescaled. If TRUE, the stomach occurences are upscaled by dividing them by the maximum occurrences / the number of non-empty stomach.

literature_diets

A dataframe containing the diet proportions found in the literature in a format similar to the stomach content data: the first row contains the names of the prey trophic groups, the headers contains the names of the consumer / predator trophic groups, and the rest are the average proportions of this prey in the predator's diet according to a literature study. The last row contains the pedigree score associated to the literature findings for each predators, a number between 0 and 1 indicating how much the literature findings are relevant estimates for the input data.

nb_literature

A vector of one number containing the equivalent number of stomach for the literature priors on the eta variable

literature_slope

A vector of one number containing the slope of the linear relationship between the pedigree scores and the PIs' coefficients of variation (CVs)

Examples

example_biotracer_data <- read.csv(system.file("extdata", "example_biotracer_data.csv",
                                               package = "EcoDiet"))
example_stomach_data <- read.csv(system.file("extdata", "example_stomach_data.csv",
                                             package = "EcoDiet"))

data <- preprocess_data(biotracer_data = example_biotracer_data,
                        trophic_discrimination_factor = c(0.8, 3.4),
                        literature_configuration = FALSE,
                        stomach_data = example_stomach_data)
                        
example_literature_diets <- read.csv(system.file("extdata", "example_literature_diets.csv",
                                                 package = "EcoDiet"))
                        
data2 <- preprocess_data(biotracer_data = example_biotracer_data,
                         trophic_discrimination_factor = c(0.8, 3.4),
                         literature_configuration = TRUE,
                         stomach_data = example_stomach_data,
                         literature_diets = example_literature_diets,
                         nb_literature = 10,
                         literature_slope = 0.5)