BBMBC: Beta-Binomial Models for Evaluating Overdispersed B-Cell Phenotyping Data

Provides tools for statistical power calculations to plan early-phase vaccine and immunotherapy trials comparing rare, antigen-specific immune cell frequencies. Originally designed for memory B-cell phenotyping studies and generalizable to high-parameter flow cytometry assays across vaccinology and immuno-oncology, the models overdispersed count distributions using Beta-Binomial generalized linear models. Unlike rank-based tests, this approach retains response magnitude information while accounting for extra-binomial variation, small sample sizes, unequal between-participant variability, and uncertainty in pilot estimates. Includes method-of-moments dispersion estimation, likelihood ratio testing across experimental arms, and Monte Carlo simulation to evaluate statistical power with well-controlled empirical Type I error rates. This workflow is directly motivated by the analysis of immunology data in "Vaccination with mRNA-encoded membrane-anchored HIV envelope trimers elicited tier 2 neutralizing antibodies in a phase 1 clinical trial" (Parks et al. 2025) <doi:10.1126/scitranslmed.ady6831>.

Version: 1.0.3
Depends: R (≥ 4.1.0)
Imports: doParallel, foreach, glmmTMB, parallel, stats, utils
Suggests: testthat (≥ 3.0.0)
Published: 2026-11-22
DOI: 10.32614/CRAN.package.BBMBC
Author: Andrew Shin [aut, cre]
Maintainer: Andrew Shin <ashin2 at uw.edu>
License: MIT + file LICENSE
NeedsCompilation: no
Materials: NEWS
CRAN checks: BBMBC results

Documentation:

Reference manual: BBMBC.html , BBMBC.pdf

Downloads:

Package source: BBMBC_1.0.3.tar.gz
Windows binaries: r-devel: BBMBC_1.0.3.zip, r-release: BBMBC_1.0.1.zip, r-oldrel: BBMBC_1.0.1.zip
macOS binaries: r-release (arm64): BBMBC_1.0.3.tgz, r-oldrel (arm64): BBMBC_1.0.3.tgz, r-release (x86_64): BBMBC_1.0.3.tgz, r-oldrel (x86_64): BBMBC_1.0.3.tgz
Old sources: BBMBC archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=BBMBC to link to this page.