Package: mbmm
Type: Package
Title: Mechanistic Bayesian Mixture Models for Latent Immunological
        States
Version: 0.1.0
URL: https://linhdo154.github.io/MBMM/, https://github.com/linhdo154/MBMM
BugReports: https://github.com/linhdo154/MBMM/issues
Authors@R: c(
    person(
      given = "Linh",
      family = "Do",
      email = "thuylinh.do@duke.edu",
      role = c("aut", "cre")
    ),
    person(
      given = "Janice",
      family = "McCarthy",
      role = "aut"
    ),
    person(
      given = "Bruce",
      family = "Rogers",
      email = "bruce.rogers2@duke.edu",
      role = "aut"
    )
  )
Description: A general implementation of the Mechanistic Bayesian Mixture
    Model (MBMM) framework of Do, McCarthy, and Rogers (2026), "A Mechanistic
    Bayesian Framework for Latent-State Inference in Vaccine and Infection
    Immunology". MBMM infers posterior probabilities of a latent
    immunological state (e.g. productive infection vs. protection/transient
    detection, or a boost-responsive vs. weak/stable antibody phenotype)
    from sparse longitudinal biomarker trajectories, by embedding an
    application-specific ODE trajectory model inside a Bayesian mixture
    likelihood evaluated against Approximate Bayesian Computation (ABC)
    reference banks, and linking covariates to latent-state probabilities
    through a logistic (K = 2) or multinomial (K > 2) gate model. The
    package is deliberately domain-agnostic: the user supplies the
    mechanistic trajectory model (ODE simulator), the ABC prior, the
    observation model, and the covariate/gate data; the package supplies
    the shared inferential engine (ABC bank construction, marginal
    reference-bank likelihoods with leave-one-donor-out scoring,
    EM/direct-optimization gate fitting, likelihood-ratio and permutation
    testing, and fitted-threshold summaries).
License: MIT + file LICENSE
Encoding: UTF-8
Depends: R (>= 4.0)
Imports: stats, graphics, grDevices, deSolve
Suggests:
    testthat (>= 3.0.0),
    knitr,
    rmarkdown,
    ggplot2
VignetteBuilder: knitr
RoxygenNote: 7.3.3
Config/testthat/edition: 3
NeedsCompilation: no
