acmgscaler

The goal of the acmgscaler R package is to provide a robust approach for gene-level calibration of variant effect scores, such as computational predictions or functional assay results, against ACMG/AMP evidence thresholds. The package is lightweight and written entirely in base R, without additional dependencies.

Colab notebook

A plug-and-play Google Colab notebook with a simple interface is available for all users.

---

[2026-03-27] – Evidence strength calculation added to the Colab notebook (see References).

---

Installation

You can install the stable version of acmgscaler directly from R:

install.packages('acmgscaler')

# or install the development version from GitHub
# devtools::install_github('badonyi/acmgscaler')
install.packages("acmgscaler", repos = c('https://badonyi.r-universe.dev', 'https://cloud.r-project.org'))

Quick start

library(acmgscaler)
data('variant_data', package = 'acmgscaler')

# calibrate the example data
calib <- calibrate(
  df = variant_data,
  value = 'score', 
  prior = 0.1,
  group = 'gene'
)

# likelihood ratios for each variant
calib$BRCA1$likelihood_ratios

# score thresholds for ACMG/AMP evidence levels
calib$BRCA1$score_thresholds

References

@article{acmgscaler,
  title={acmgscaler: An R package and Colab for standardised gene-level variant effect score calibration within the ACMG/AMP framework},
  author={Badonyi, Mihaly and Marsh, Joseph A},
  journal={Bioinformatics},
  pages={btaf503},
  year={2025},
  publisher={Oxford University Press}
}

@article{MES,
  title={Clinical evidence yield as a framework for evaluating computational predictors and multiplexed assays of variant effect},
  author={Shang, Yifei and Badonyi, Mihaly and Marsh, Joseph A},
  journal={bioRxiv},
  year={2026},
  publisher={Cold Spring Harbor Laboratory}
}