MalReBay

MalReBay is an R package for Bayesian molecular correction of malaria therapeutic efficacy studies (TES). Given paired genotyping data from a patient’s Day 0 infection and a later recurrence, it estimates the posterior probability of recrudescence (treatment failure) versus reinfection (new parasite). It supports length-polymorphic markers (microsatellites, MSP, GLURP) and amplicon sequencing data, and uses Stan HMC-NUTS via cmdstanr for inference.

Installation

Installing MalReBay takes three short steps and about 10 minutes in total, almost all of it hands-off waiting. You only need to do this once per computer — after that, library(MalReBay) is all you need, in every future R session.

Why so many steps? MalReBay’s statistical engine is written in Stan, a language for Bayesian modelling that needs to be compiled into fast machine code. Step 1 installs that compiler toolkit (CmdStan) and step 2 checks it’s ready to go. Step 3 installs MalReBay itself, which compiles its model against the CmdStan from step 1 — that’s why the order below matters: CmdStan has to be in place before you install MalReBay.

Prerequisites

• R version 4.1.0 or later. Check yours by running R.version.string in the R console.
• A C++ compiler, needed to build CmdStan and MalReBay’s model. This is usually a one-time, one-command install:
  • Windows: install Rtools (pick the version matching your R version).
  • macOS: open Terminal and run xcode-select --install.
  • Linux (Debian/Ubuntu): open a terminal and run sudo apt-get install build-essential.

Step 1 — Install CmdStan

# install.packages("cmdstanr", repos = c("https://stan-dev.r-universe.dev", getOption("repos")))
cmdstanr::install_cmdstan()

This downloads and compiles CmdStan, MalReBay’s statistical engine — it takes around 5 minutes and shows its own progress as it goes.

Step 2 — Check your setup (optional but recommended)

cmdstanr::check_cmdstan_toolchain()

This confirms your C++ compiler is correctly set up before you try to install MalReBay, so any problem is caught here with a clear message rather than a confusing error later. If it reports a problem, follow its instructions (usually pointing back to the Prerequisites step above), then run it again.

Step 3 — Install MalReBay

# install.packages("remotes")  # run this line first if you don't have remotes
remotes::install_github("SwissTPH/MalReBay")

This step compiles MalReBay’s Stan model against the CmdStan from step 1, so it takes a little longer than a typical package install (roughly a minute) — that one-time cost is also why the order matters: if CmdStan isn’t installed yet, this step will fail.

That’s it! Open a fresh R session and confirm everything worked:

library(MalReBay)

If this loads without a NOTE: CmdStan is not installed... message, you’re ready to go — see Quick Start below.

Troubleshooting

• “unused argument” or a compiler error during step 3: re-run cmdstanr::check_cmdstan_toolchain() from step 2 — it usually points directly at the missing piece (most often a missing C++ compiler).
• Installed once already, now reinstalling after making changes to the package: every reinstall recompiles the Stan model from scratch, so step 3 will always take about a minute, even for small changes elsewhere in the package.
• Still stuck? Open an issue at https://github.com/SwissTPH/MalReBay/issues with the exact error message — it’s the fastest way for us to help.

Quick Start

Try MalReBay right away on its bundled example dataset — no files of your own needed yet:

library(MalReBay)

results <- MalReBay()

head(results$posterior_probabilities)

Once you’re ready to use your own data, point MalReBay() at your files:

results <- MalReBay(
  filepath        = "path/to/genotype_data.xlsx",
  marker_filepath = "path/to/marker_info.xlsx",
  output_folder   = "my_results"
)

head(results$posterior_probabilities)

For a full walkthrough of the input file formats and a worked example, see the package vignette.