# openpharma/mmrm architecture

> Architecture diagram of the GitHub repository openpharma/mmrm, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-07). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/openpharma/mmrm
- Repository: https://github.com/openpharma/mmrm

## Overview

mmrm is an R package for fitting marginal mixed models for repeated-measures continuous outcomes. The main workflow is to supply a formula and longitudinal data to the fitting interface, estimate fixed effects and covariance parameters with ML or REML, then summarize, test contrasts, or obtain estimated means and predictions. Covariance construction, cached factorizations and derivatives, and C++ likelihood calculations form the computational core. The README documents additional inference and ecosystem capabilities; sampled source excerpts establish the C++ component relationships, but do not expose all R-level dispatch wiring.

Main entry points: [`R/mmrm-methods.R`](https://github.com/openpharma/mmrm/blob/main/R/mmrm-methods.R) (Fit methods), [`R/fit.R`](https://github.com/openpharma/mmrm/blob/main/R/fit.R) (Fit interface), [`src/mmrm.cpp`](https://github.com/openpharma/mmrm/blob/main/src/mmrm.cpp) (Likelihood fitting), [`src/chol_cache.h`](https://github.com/openpharma/mmrm/blob/main/src/chol_cache.h) (Cholesky caches)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_model["Model fitting"]
  node_fit_api["Fit interface<br/>[fit.R]"]
  node_formula["Formula utilities<br/>[utils-formula.R]"]
  node_cov_spec["Covariance specification<br/>[cov_struct.R]"]
  node_fit_methods["Fit methods<br/>[mmrm-methods.R]"]
end

subgraph group_compute["Numerical backend"]
  node_covariance["Covariance matrices<br/>[covariance.h]"]
  node_chol_cache["Cholesky caches<br/>[chol_cache.h]"]
  node_derivatives["Covariance derivatives<br/>[derivatives.h]"]
  node_likelihood["Likelihood fitting<br/>[mmrm.cpp]"]
  node_tmb_bridge["TMB interface<br/>[tmb.R]"]
  node_kr_components["KR components<br/>[kr_comp.cpp]"]
end

subgraph group_inference["Inference and outputs"]
  node_empirical["Empirical covariance<br/>[empirical.R]"]
  node_testing["Contrast testing<br/>[testing.R]"]
  node_satterthwaite["Satterthwaite inference<br/>[satterthwaite.R]"]
  node_kenward["Kenward-Roger inference<br/>[kenwardroger.R]"]
  node_means["Estimated means<br/>[interop-emmeans.R]"]
  node_prediction["Prediction<br/>[predict.cpp]"]
  node_residuals["Residual diagnostics<br/>[residual.R]"]
  node_tidy["Tidy outputs<br/>[tidiers.R]"]
end

subgraph group_integrations["Ecosystem integrations"]
  node_car["car integration<br/>[interop-car.R]"]
  node_parsnip["parsnip engine<br/>[interop-parsnip.R]"]
  node_emmeans_ext["emmeans support<br/>[interop-emmeans.R]"]
end

node_analyst(("Analyst"))
node_user(("Model consumer"))

node_analyst -->|"supplies model"| node_fit_api
node_fit_api -.->|"uses formula"| node_formula
node_fit_api -.->|"specifies covariance"| node_cov_spec
node_fit_api -.->|"fits through"| node_tmb_bridge
node_tmb_bridge -.->|"evaluates likelihood"| node_likelihood
node_likelihood -->|"gets factors"| node_chol_cache
node_likelihood -->|"uses covariance"| node_covariance
node_chol_cache -->|"builds factors"| node_covariance
node_fit_api -.->|"returns fit"| node_fit_methods
node_fit_methods -.->|"summarizes with"| node_satterthwaite
node_fit_methods -.->|"supports robust inference"| node_empirical
node_fit_methods -.->|"tests contrasts"| node_testing
node_testing -.->|"adjusts degrees freedom"| node_kenward
node_kenward -.->|"computes components"| node_kr_components
node_kr_components -->|"gets derivatives"| node_derivatives
node_derivatives -->|"extends cache"| node_chol_cache
node_fit_methods -.->|"supports means"| node_means
node_fit_methods -.->|"predicts outcomes"| node_prediction
node_fit_methods -.->|"extracts residuals"| node_residuals
node_fit_methods -.->|"formats results"| node_tidy
node_fit_methods -.->|"integrates with"| node_car
node_fit_methods -.->|"registers engine"| node_parsnip
node_means -.->|"uses support"| node_emmeans_ext
node_user -.->|"consumes results"| node_fit_methods

click node_fit_api "https://github.com/openpharma/mmrm/blob/main/R/fit.R"
click node_formula "https://github.com/openpharma/mmrm/blob/main/R/utils-formula.R"
click node_cov_spec "https://github.com/openpharma/mmrm/blob/main/R/cov_struct.R"
click node_covariance "https://github.com/openpharma/mmrm/blob/main/src/covariance.h"
click node_chol_cache "https://github.com/openpharma/mmrm/blob/main/src/chol_cache.h"
click node_derivatives "https://github.com/openpharma/mmrm/blob/main/src/derivatives.h"
click node_likelihood "https://github.com/openpharma/mmrm/blob/main/src/mmrm.cpp"
click node_tmb_bridge "https://github.com/openpharma/mmrm/blob/main/R/tmb.R"
click node_fit_methods "https://github.com/openpharma/mmrm/blob/main/R/mmrm-methods.R"
click node_empirical "https://github.com/openpharma/mmrm/blob/main/R/empirical.R"
click node_testing "https://github.com/openpharma/mmrm/blob/main/R/testing.R"
click node_satterthwaite "https://github.com/openpharma/mmrm/blob/main/R/satterthwaite.R"
click node_kenward "https://github.com/openpharma/mmrm/blob/main/R/kenwardroger.R"
click node_kr_components "https://github.com/openpharma/mmrm/blob/main/src/kr_comp.cpp"
click node_means "https://github.com/openpharma/mmrm/blob/main/R/interop-emmeans.R"
click node_prediction "https://github.com/openpharma/mmrm/blob/main/src/predict.cpp"
click node_residuals "https://github.com/openpharma/mmrm/blob/main/R/residual.R"
click node_tidy "https://github.com/openpharma/mmrm/blob/main/R/tidiers.R"
click node_car "https://github.com/openpharma/mmrm/blob/main/R/interop-car.R"
click node_parsnip "https://github.com/openpharma/mmrm/blob/main/R/interop-parsnip.R"
click node_emmeans_ext "https://github.com/openpharma/mmrm/blob/main/R/interop-emmeans.R"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_fit_api,node_formula,node_cov_spec,node_fit_methods toneBlue
class node_covariance,node_chol_cache,node_derivatives,node_likelihood,node_tmb_bridge,node_kr_components toneAmber
class node_empirical,node_testing,node_satterthwaite,node_kenward,node_means,node_prediction,node_residuals,node_tidy toneMint
class node_car,node_parsnip,node_emmeans_ext toneRose
class node_analyst,node_user toneIndigo
```

## Components

### Model fitting

- [Fit interface](https://github.com/openpharma/mmrm/blob/main/R/fit.R): `R/fit.R`
- [Formula utilities](https://github.com/openpharma/mmrm/blob/main/R/utils-formula.R): `R/utils-formula.R`
- [Covariance specification](https://github.com/openpharma/mmrm/blob/main/R/cov_struct.R): `R/cov_struct.R`
- [Fit methods](https://github.com/openpharma/mmrm/blob/main/R/mmrm-methods.R): `R/mmrm-methods.R`

### Numerical backend

- [Covariance matrices](https://github.com/openpharma/mmrm/blob/main/src/covariance.h): `src/covariance.h`
- [Cholesky caches](https://github.com/openpharma/mmrm/blob/main/src/chol_cache.h): `src/chol_cache.h`
- [Covariance derivatives](https://github.com/openpharma/mmrm/blob/main/src/derivatives.h): `src/derivatives.h`
- [Likelihood fitting](https://github.com/openpharma/mmrm/blob/main/src/mmrm.cpp): `src/mmrm.cpp`
- [TMB interface](https://github.com/openpharma/mmrm/blob/main/R/tmb.R): `R/tmb.R`
- [KR components](https://github.com/openpharma/mmrm/blob/main/src/kr_comp.cpp): `src/kr_comp.cpp`

### Inference and outputs

- [Empirical covariance](https://github.com/openpharma/mmrm/blob/main/R/empirical.R): `R/empirical.R`
- [Contrast testing](https://github.com/openpharma/mmrm/blob/main/R/testing.R): `R/testing.R`
- [Satterthwaite inference](https://github.com/openpharma/mmrm/blob/main/R/satterthwaite.R): `R/satterthwaite.R`
- [Kenward-Roger inference](https://github.com/openpharma/mmrm/blob/main/R/kenwardroger.R): `R/kenwardroger.R`
- [Estimated means](https://github.com/openpharma/mmrm/blob/main/R/interop-emmeans.R): `R/interop-emmeans.R`
- [Prediction](https://github.com/openpharma/mmrm/blob/main/src/predict.cpp): `src/predict.cpp`
- [Residual diagnostics](https://github.com/openpharma/mmrm/blob/main/R/residual.R): `R/residual.R`
- [Tidy outputs](https://github.com/openpharma/mmrm/blob/main/R/tidiers.R): `R/tidiers.R`

### Ecosystem integrations

- [car integration](https://github.com/openpharma/mmrm/blob/main/R/interop-car.R): `R/interop-car.R`
- [parsnip engine](https://github.com/openpharma/mmrm/blob/main/R/interop-parsnip.R): `R/interop-parsnip.R`
- [emmeans support](https://github.com/openpharma/mmrm/blob/main/R/interop-emmeans.R): `R/interop-emmeans.R`

### Other components

- **Analyst**
- **Model consumer**

## Connections

- Analyst → Fit interface: supplies model (evidence: `README.md`)
- Fit interface → Formula utilities: uses formula
- Fit interface → Covariance specification: specifies covariance
- Fit interface → TMB interface: fits through
- TMB interface → Likelihood fitting: evaluates likelihood
- Likelihood fitting → Cholesky caches: gets factors (evidence: `src/mmrm.cpp`)
- Likelihood fitting → Covariance matrices: uses covariance (evidence: `src/mmrm.cpp`)
- Cholesky caches → Covariance matrices: builds factors (evidence: `src/chol_cache.h`)
- Fit interface → Fit methods: returns fit
- Fit methods → Satterthwaite inference: summarizes with
- Fit methods → Empirical covariance: supports robust inference
- Fit methods → Contrast testing: tests contrasts
- Contrast testing → Kenward-Roger inference: adjusts degrees freedom
- Kenward-Roger inference → KR components: computes components
- KR components → Covariance derivatives: gets derivatives
- Covariance derivatives → Cholesky caches: extends cache (evidence: `src/derivatives.h`)
- Fit methods → Estimated means: supports means (evidence: `README.md`)
- Fit methods → Prediction: predicts outcomes
- Fit methods → Residual diagnostics: extracts residuals
- Fit methods → Tidy outputs: formats results
- Fit methods → car integration: integrates with (evidence: `README.md`)
- Fit methods → parsnip engine: registers engine (evidence: `README.md`)
- Estimated means → emmeans support: uses support (evidence: `README.md`)
- Model consumer → Fit methods: consumes results
