Skip to contents

First-stage plot values from an analysis

Source: R/breedbase_analysis.R
first_stage_values.Rd

Turns analyses of a trial into one value per field plot, ready for save_breedbase_first_stage(): the genotype's adjusted mean plus the plot's residual. This removes the design effects (blocks, rows, columns, replicates) and the spatial trend, and keeps the plot-to-plot variation, so the result has as many plots as the field trial. With SpATS (spatial = TRUE or stage1 = "spatial") these are spatially corrected plot values.

Usage

first_stage_values(
  fits,
  design = NULL,
  trial_name = NULL,
  trait = NULL,
  plot_name_pattern = "{trial}_{accession}_{block}_{rep}",
  plot_names = NULL,
  breedbase_plots = NULL
)

Arguments

fits

One analysis, or a named list of analyses, one per trait (the names are the Breedbase trait names). Analyses from analyze_augmented_row_column_design(), analyze_augmented_prep_design(), analyze_augmented_prep_two_stage() (stage-1 fits per location) or analyze_resolvable_row_column().

design

The design that was saved to Breedbase. Leave it NULL for data read from Breedbase (breedbase_phenotypes()): the plot names are then taken from the data.

trial_name

The trial name used when saving the design (not needed without a design).

trait

Trait name when fits is a single analysis.

plot_name_pattern, plot_names

The plot naming used in save_breedbase().

breedbase_plots

Optional plots of the saved trial, from breedbase_trial_plots(). The design is then checked against the saved trial (the same accession at every plot number), and the plot names are taken exactly as saved.

Value

A tibble with plot_name, trial_name, accession, plot_number, block, rep, row, col, is_check and one column per trait. Its attribute "method" is "spatial" when any analysis used SpATS, otherwise "mixed".

Examples

rc <- augmented_row_column_design(paste0("G", 1:108), paste0("C", 1:4),
                                  12, 12, 4, 4, n_candidates = 5, seed = 1)
d <- rc$design
set.seed(1)
d$yield <- 25 + 0.4 * d$row + rnorm(nrow(d), sd = 2)
fit <- analyze_augmented_row_column_design(d, "yield")
values <- first_stage_values(fit, rc, "AYT_2026_Ibadan", trait = "fresh root yield")
head(values)
#> # A tibble: 6 × 10
#>   plot_name    trial_name accession plot_number block   rep   row   col is_check
#>   <chr>        <chr>      <chr>           <int> <int> <int> <int> <int> <lgl>   
#> 1 AYT_2026_Ib… AYT_2026_… G51                 1     1     1     1     1 FALSE   
#> 2 AYT_2026_Ib… AYT_2026_… G64                 2     1     1     1     2 FALSE   
#> 3 AYT_2026_Ib… AYT_2026_… G97                 3     1     1     1     3 FALSE   
#> 4 AYT_2026_Ib… AYT_2026_… C3                  4     1     1     1     4 TRUE    
#> 5 AYT_2026_Ib… AYT_2026_… G8                  5     2     1     1     5 FALSE   
#> 6 AYT_2026_Ib… AYT_2026_… C2                  6     2     2     1     6 TRUE    
#> # ℹ 1 more variable: `fresh root yield` <dbl>