This book provides examples of exposure-response analysis with Bayesian methods.

Install necessary packages

The examples utilizes BayesERtools package.

• Tutorial (BayesERbook): https://genentech.github.io/BayesERbook/
• Package documentation: https://genentech.github.io/BayesERtools/
• GitHub repo of the package: https://github.com/genentech/BayesERtools/

You can install the package as follows:

install.packages('BayesERtools')
# devtools::install_github("genentech/BayesERtools") # development version

Quick Example

library(BayesERtools)
library(dplyr)
library(ggplot2)
theme_set(theme_bw(base_size = 12))

# Data
data(d_sim_binom_cov)
df_er_ae_hgly2 <-
  d_sim_binom_cov |>
  mutate(AUCss_1000 = AUCss / 1000) |>
  filter(AETYPE == "hgly2")

# Fit a model
ermod <- dev_ermod_bin(
  data = df_er_ae_hgly2,
  var_resp = "AEFLAG",
  var_exposure = "AUCss_1000"
)

# Goodness-of-fit plot
plot_er_gof(ermod, var_group = "Dose_mg", show_coef_exp = TRUE) *
  coord_transform(x = "log10") *
  scale_x_continuous(
    breaks = xgxr::xgx_breaks_log10,
    minor_breaks = xgxr::xgx_minor_breaks_log10
  )

Model types supported by BayesERtools

	Binary endpoint		Continuous endpoint
	Linear (logit)	Emax (logit)	Linear	Emax
backend	rstanarm	rstanemax	rstanarm	rstanemax
reference	🔗	🔗	🔗	🔗
develop model	✅	✅	✅	✅
simulate & plot ER	✅	✅	✅	✅
exposure metrics selection	✅	✅	✅	✅
covariate selection	✅	❌	✅	❌
covariate forest plot	✅	❌	✅	❌
✅ Available, 🟡 In plan/under development, ❌ Not in a current plan

Note for developer

Run usethis::use_tidy_style(strict = FALSE) before committing to ensure that the code is formatted appropriately.