Reorganize repo into package-ready structure with pkgdown + mock data + tests

.Rbuildignore

@@ -8,3 +8,12 @@
 ^vignettes/articles/*_files$
 ^vignettes/articles$
 ^codecov\.yml$
+^data-raw$
+^README\.Rmd$
+^\.lintr\.R$
+^_quarto\.yml$
+^.*\.qmd$
+^.*\.png$
+^Vary.*\.r$
+^sim_.*\.r$
+^simulation.*\.qmd$

.github/workflows/R-CMD-check.yaml

@@ -0,0 +1,48 @@
+on:
+  push:
+    branches: [main, master]
+  pull_request:
+    branches: [main, master]
+
+name: R-CMD-check
+
+permissions:
+  contents: read
+
+jobs:
+  R-CMD-check:
+    runs-on: ${{ matrix.config.os }}
+
+    name: ${{ matrix.config.os }} (${{ matrix.config.r }})
+
+    strategy:
+      fail-fast: false
+      matrix:
+        config:
+          - {os: ubuntu-latest, r: 'release'}
+          - {os: ubuntu-latest, r: 'devel', http-user-agent: 'release'}
+
+    env:
+      GITHUB_PAT: ${{ github.token }}
+      R_KEEP_PKG_SOURCE: yes
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: r-lib/actions/setup-pandoc@v2
+
+      - uses: r-lib/actions/setup-r@v2
+        with:
+          r-version: ${{ matrix.config.r }}
+          http-user-agent: ${{ matrix.config.http-user-agent }}
+          use-public-rspm: true
+
+      - uses: r-lib/actions/setup-r-dependencies@v2
+        with:
+          extra-packages: any::rcmdcheck
+          needs: check
+
+      - uses: r-lib/actions/check-r-package@v2
+        with:
+          upload-snapshots: true
+          build_args: 'c("--no-manual","--compact-vignettes=gs+qpdf")'

.github/workflows/pkgdown.yaml

@@ -0,0 +1,46 @@
+on:
+  push:
+    branches: [main, master]
+  release:
+    types: [published]
+  workflow_dispatch:
+
+name: pkgdown
+
+permissions:
+  contents: write
+  pages: write
+
+jobs:
+  pkgdown:
+    runs-on: ubuntu-latest
+    # Only restrict concurrency for non-PR jobs
+    concurrency:
+      group: pkgdown-${{ github.event_name != 'pull_request' || github.run_id }}
+    env:
+      GITHUB_PAT: ${{ github.token }}
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: r-lib/actions/setup-pandoc@v2
+
+      - uses: r-lib/actions/setup-r@v2
+        with:
+          use-public-rspm: true
+
+      - uses: r-lib/actions/setup-r-dependencies@v2
+        with:
+          extra-packages: any::pkgdown, local::.
+          needs: website
+
+      - name: Build site
+        run: pkgdown::build_site_github_pages(new_process = FALSE, install = FALSE)
+        shell: Rscript {0}
+
+      - name: Deploy to GitHub pages 🚀
+        if: github.event_name != 'pull_request'
+        uses: JamesIves/github-pages-deploy-action@v4
+        with:
+          clean: false
+          branch: gh-pages
+          folder: docs

.gitignore

@@ -8,7 +8,10 @@ simulation1_edited_files
 simulation1_edited.html
 *.rmarkdown
 **/.quarto/
-data/
+data/*
+!data/.gitignore
+!data/mock_posterior_draws.rda
+!data/mock_case_data.rda
 /.quarto/
 *.pdf
 README_files

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: shigella
 Title: What the Package Does (One Line, Title Case)
-Version: 0.0.0.9004
+Version: 0.0.0.9005
 Authors@R: c(
     person("Kwan Ho", "Lee", , "ksjlee@ucdavis.edu", role = c("aut", "cre")),
     person("Douglas Ezra", "Morrison", , "demorrison@ucdavis.edu", role = c("aut"),
@@ -9,14 +9,15 @@ Description: What the package does (one paragraph).
 License: MIT + file LICENSE
 Encoding: UTF-8
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.3.2
+RoxygenNote: 7.3.3
 Imports: 
+    cli,
     dplyr,
-    future,
-    future.apply,
     ggplot2,
-    gridExtra,
-    serodynamics (>= 0.0.0.9011)
+    rlang,
+    serocalculator,
+    tibble,
+    tidyr
 URL: https://ucd-serg.github.io/shigella/
 Remotes: 
     UCD-SERG/serodynamics
@@ -26,8 +27,11 @@ Suggests:
     coda,
     forcats,
     furrr,
+    future,
+    future.apply,
     ggeasy,
     ggmcmc,
+    gridExtra,
     gtsummary,
     haven,
     here,
@@ -42,15 +46,18 @@ Suggests:
     purrr,
     quarto,
     readxl,
+    rmarkdown,
     runjags,
     scales,
+    serodynamics (>= 0.0.0.9011),
     spelling,
     table1,
-    tibble,
-    tidyr,
+    testthat (>= 3.0.0),
     tidyverse
+VignetteBuilder: knitr
 Depends: 
     R (>= 3.5)
 LazyData: true
+Config/testthat/edition: 3
 Config/Needs/website: quarto
 Language: en-US

LICENSE

@@ -0,0 +1,2 @@
+YEAR: 2024
+COPYRIGHT HOLDER: Kwan Ho Lee, Douglas Ezra Morrison

NAMESPACE

@@ -1,6 +1,12 @@
 # Generated by roxygen2: do not edit by hand
 
-export(postprocess_jags_output)
+export(compute_residual_metrics)
+export(fig2_overall_newperson)
+export(fmt_mci)
+export(make_model_comparison_table)
+export(model_comparison_table)
+export(prep_newperson_params)
+export(process_shigella_data)
 importFrom(dplyr,any_of)
 importFrom(dplyr,filter)
 importFrom(dplyr,mutate)
@@ -13,3 +19,5 @@ importFrom(ggplot2,geom_point)
 importFrom(ggplot2,ggplot)
 importFrom(ggplot2,labs)
 importFrom(ggplot2,theme_minimal)
+importFrom(rlang,.data)
+importFrom(rlang,ensym)

NEWS.md

@@ -10,7 +10,7 @@
 
 * added `.lintr` configuration file
 * Moved helper functions into `R/` subdirectory (#1)
-* simplified spell-check Github Action (#3)
+* simplified spell-check GitHub Action (#3)
 
 # shigella 0.0.0
 

R/compute_residual_metrics.R

@@ -0,0 +1,183 @@
+#' Residual-based posterior predictive metrics for longitudinal antibody curves
+#'
+#' Computes residuals between observed antibody measurements and posterior
+#' median predictions evaluated at the observed time points. Returns pointwise
+#' residuals or aggregated error metrics (MAE, RMSE, SSE) at multiple
+#' summary levels.
+#'
+#' @param model A data frame of posterior draws in long format with columns:
+#'   \code{Subject}, \code{Iso_type}, \code{Chain}, \code{Iteration},
+#'   \code{Parameter}, \code{value}.
+#' @param dataset A \code{serodynamics} case dataset produced by
+#'   \code{serodynamics::as_case_data()} (must contain \code{id} and
+#'   \code{antigen_iso} columns, and time/value attributes).
+#' @param ids Character vector of subject IDs to include (matched against
+#'   \code{dataset$id} and \code{model$Subject}).
+#' @param antigen_iso Character scalar specifying the antigen/isotype to analyze
+#'   (matched against \code{dataset$antigen_iso} and \code{model$Iso_type}).
+#' @param scale Scale on which to compute residuals. One of \code{"original"}
+#'   or \code{"log"}. If \code{"log"}, residuals are computed on the natural
+#'   log scale and observations/predictions \eqn{\le 0} are removed.
+#' @param summary_level Level at which to summarize metrics. One of:
+#'   \describe{
+#'     \item{\code{"pointwise"}}{
+#'       Return pointwise residuals at each observed time point.
+#'     }
+#'     \item{\code{"id_antigen"}}{
+#'       Summarize by \code{id} and \code{antigen_iso}.
+#'     }
+#'     \item{\code{"antigen"}}{Summarize by \code{antigen_iso} only.}
+#'     \item{\code{"overall"}}{Single summary across all included observations.}
+#'   }
+#'
+#' @return A tibble. For \code{summary_level = "pointwise"}, returns
+#'   per-observation residuals. Otherwise returns MAE, RMSE, SSE, and
+#'   \code{n_obs} at the requested summary level.
+#'
+#' @details
+#' The posterior predictive summary uses the posterior median at each observed
+#' time point, with a 95\% credible interval computed from draw-level
+#' predictions. Predictions are generated by
+#' \code{\link{predict_posterior_at_times}}.
+#'
+#' @examples
+#' \dontrun{
+#' # Per-ID error metrics (original scale)
+#' m_id <- compute_residual_metrics(
+#'   model = overall_sf2a,
+#'   dataset = dL_clean_sf2a,
+#'   ids = unique(dL_clean_sf2a$id),
+#'   antigen_iso = "IgG",
+#'   scale = "original",
+#'   summary_level = "id_antigen"
+#' )
+#'
+#' # Pointwise residuals on log scale
+#' r_pw <- compute_residual_metrics(
+#'   model = overall_sf2a,
+#'   dataset = dL_clean_sf2a,
+#'   ids = "SOSAR-22008",
+#'   antigen_iso = "IgA",
+#'   scale = "log",
+#'   summary_level = "pointwise"
+#' )
+#' }
+#'
+#' @export
+compute_residual_metrics <- function(model,
+                                     dataset,
+                                     ids,
+                                     antigen_iso,
+                                     scale = c("original", "log"),
+                                     summary_level = c(
+                                       "id_antigen",
+                                       "pointwise",
+                                       "antigen",
+                                       "overall"
+                                     )) {
+
+  scale <- match.arg(scale)
+  summary_level <- match.arg(summary_level)
+
+  time_var  <- get_timeindays_var(dataset)
+  value_var <- serocalculator::get_values_var(dataset)
+
+  observed_data <- dataset |>
+    dplyr::rename(
+      t   = !!rlang::sym(time_var),
+      obs = !!rlang::sym(value_var)
+    ) |>
+    dplyr::select(dplyr::all_of(c("id", "t", "obs", "antigen_iso"))) |>
+    dplyr::mutate(id = as.character(.data$id)) |>
+    dplyr::filter(.data$id %in% ids, .data$antigen_iso == antigen_iso)
+
+  if (nrow(observed_data) == 0) {
+    cli::cli_abort(
+      "No observed data found for the specified IDs and antigen_iso."
+    )
+  }
+
+  obs_times <- sort(unique(observed_data$t))
+
+  predictions_all <- predict_posterior_at_times(
+    model = model,
+    ids = ids,
+    antigen_iso = antigen_iso,
+    times = obs_times
+  )
+
+  pred_summary <- predictions_all |>
+    dplyr::summarise(
+      .by = c("id", "t"),
+      pred_med   = stats::median(.data$res, na.rm = TRUE),
+      pred_lower = stats::quantile(.data$res, probs = 0.025, na.rm = TRUE),
+      pred_upper = stats::quantile(.data$res, probs = 0.975, na.rm = TRUE)
+    ) |>
+    dplyr::mutate(id = as.character(.data$id))
+
+  residual_data <- observed_data |>
+    dplyr::inner_join(pred_summary, by = c("id", "t"))
+
+  if (scale == "log") {
+    residual_data <- residual_data |>
+      dplyr::filter(.data$obs > 0, .data$pred_med > 0) |>
+      dplyr::mutate(
+        residual     = log(.data$obs) - log(.data$pred_med),
+        abs_residual = abs(.data$residual),
+        sq_residual  = .data$residual^2
+      )
+  } else {
+    residual_data <- residual_data |>
+      dplyr::mutate(
+        residual     = .data$obs - .data$pred_med,
+        abs_residual = abs(.data$residual),
+        sq_residual  = .data$residual^2
+      )
+  }
+
+  if (summary_level == "pointwise") {
+    return(residual_data |>
+      dplyr::select(
+        "id",
+        "antigen_iso",
+        "t",
+        "obs",
+        "pred_med",
+        "pred_lower",
+        "pred_upper",
+        "residual",
+        "abs_residual",
+        "sq_residual"
+      ))
+  }
+
+  if (summary_level == "id_antigen") {
+    return(residual_data |>
+      dplyr::summarise(
+        .by = c("id", "antigen_iso"),
+        MAE = mean(.data$abs_residual, na.rm = TRUE),
+        RMSE = sqrt(mean(.data$sq_residual, na.rm = TRUE)),
+        SSE = sum(.data$sq_residual, na.rm = TRUE),
+        n_obs = dplyr::n()
+      ))
+  }
+
+  if (summary_level == "antigen") {
+    return(residual_data |>
+      dplyr::summarise(
+        .by = "antigen_iso",
+        MAE = mean(.data$abs_residual, na.rm = TRUE),
+        RMSE = sqrt(mean(.data$sq_residual, na.rm = TRUE)),
+        SSE = sum(.data$sq_residual, na.rm = TRUE),
+        n_obs = dplyr::n()
+      ))
+  }
+
+  residual_data |>
+    dplyr::summarise(
+      MAE = mean(.data$abs_residual, na.rm = TRUE),
+      RMSE = sqrt(mean(.data$sq_residual, na.rm = TRUE)),
+      SSE = sum(.data$sq_residual, na.rm = TRUE),
+      n_obs = dplyr::n()
+    )
+}