Plot fitted dose-response curves for multiple isolates

`plot_EC50_curves()` plots an object returned by [estimate_EC50()] or [ec50_multimodel()]. It uses the formula, original data, grouping columns, and fitted `drc` models stored in that result, so users do not need to repeat the modeling arguments. For compatibility, the function also accepts the original formula/data interface.

Usage

plot_EC50_curves(
  x,
  data = NULL,
  isolate_col = NULL,
  strata_col = NULL,
  fct = NULL,
  color_col = NULL,
  facet_col = NULL,
  facet_row = NULL,
  models = "all",
  n_points = 200,
  log_x = TRUE,
  point_size = 2,
  point_alpha = 0.8,
  line_width = 1,
  quiet = FALSE
)

Arguments

x

An object returned by [estimate_EC50()] or [ec50_multimodel()]. A two-sided formula such as `growth ~ dose` is also accepted for compatibility.

data

A data frame containing the response, dose, isolate, and optional stratification columns. Required only when `x` is a formula.

isolate_col

Character scalar naming the column that identifies each isolate. Required only when `x` is a formula.

strata_col

Optional character vector naming columns used to split the data before fitting models. Used only when `x` is a formula.

fct

A `drc` model function object such as `drc::LL.3()`, or a list of model function objects such as `list(drc::LL.3(), drc::LL.4())`. Required only when `x` is a formula.

color_col

Character scalar naming the column mapped to curve and point color. Defaults to the isolate column and is always converted to a factor before plotting.

facet_col, facet_row

Optional character scalars naming columns used for faceting. When omitted, the first two `strata_col` values are used.

models

One of `"all"`, `"best"`, or a character vector of model names to draw. `"best"` uses [best_model()] for multimodel fits.

n_points

Number of dose values used to draw each fitted curve.

log_x

Logical. If `TRUE`, use a log10 x-axis and omit non-positive dose values from the plotted data and prediction grid.

point_size, point_alpha

Size and alpha for raw observation points.

line_width

Width for fitted curves.

quiet

Logical. If `FALSE`, failed group/model fits are reported with a warning.

Value

A `ggplot2` object. The plotted curve data, observed data, and fitted models are attached to the returned object as `curve_data`, `observed_data`, and `fitted_models` attributes and list elements.

Examples

data(multi_isolate)
sample_data <- subset(
  multi_isolate,
  isolate %in% 1:4 & fungicida == "Fungicide A"
)

fit <- estimate_EC50(
  growth ~ dose,
  data = sample_data,
  isolate_col = "isolate",
  strata_col = "field",
  fct = drc::LL.3()
)
plot_EC50_curves(fit)


multi_fit <- ec50_multimodel(
  growth ~ dose,
  data = sample_data,
  isolate_col = "isolate",
  strata_col = "field",
  fct = list(drc::LL.3(), drc::LL.4())
)
plot_EC50_curves(multi_fit)