diff --git a/DESCRIPTION b/DESCRIPTION
index 784b8f4..f3ec63d 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -34,7 +34,11 @@ Imports:
     data.table,
     dplyr,
     duckdb,
+    ggplot2,
     glue,
+    grid,
+    gridExtra,
+    knitr,
     purrr,
     readr,
     reshape2,
@@ -44,7 +48,6 @@ Imports:
 biocViews: GenomeAssembly, Annotation, Sequencing
 VignetteBuilder: knitr
 Suggests:
-    knitr,
     rmarkdown,
     testthat (>= 3.0.0)
 Config/roxygen2/version: 8.0.0
diff --git a/NAMESPACE b/NAMESPACE
index ff6d425..69af5ae 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -6,6 +6,8 @@ export(CDHIT2duckdb)
 export(buildClusterFeatureMap)
 export(cleanData)
 export(cleanMetaData)
+export(generatePlots)
+export(generateSummary)
 export(prepareGenomes)
 export(proteinAnnotations2Duckdb)
 export(retrieveGenomes)
diff --git a/R/amRdataPlots.R b/R/amRdataPlots.R
new file mode 100644
index 0000000..7cf3f63
--- /dev/null
+++ b/R/amRdataPlots.R
@@ -0,0 +1,491 @@
+#' Generate a summary report for AMR metadata
+#'
+#' @param metadata_parquet Character string. Path to a Parquet file containing
+#'   standardized AMR metadata.
+#' @param out_path Character string. Directory where the Markdown report is written.
+#'
+#' @return Writes a structured, human‑readable summary report to
+#'   "<out_path>/amr_metadata_summary.md".
+#'
+#' @examples
+#' generateSummary(
+#'   metadata_parquet = "results/metadata.parquet",
+#'   out_path = "results/"
+#' )
+#'
+#' @export
+generateSummary <- function(metadata_parquet, out_path) {
+  # Little helper to apply distinct + non-empty + sorted vector
+  clean_distinct <- function(df, col) {
+    df |>
+      dplyr::distinct({{ col }}) |>
+      dplyr::filter(!is.na({{ col }}), {{ col }} != "") |>
+      dplyr::arrange({{ col }}) |>
+      dplyr::pull({{ col }})
+  }
+
+  # Format for Markdown
+  md_tbl <- function(df) {
+    knitr::kable(df, format = "pipe")
+  }
+
+  # Create a file
+  write_new <- function(path, lines) {
+    con <- file(path, open = "w", encoding = "UTF-8")
+    on.exit(close(con), add = TRUE)
+    writeLines(lines, con = con, sep = "\n", useBytes = TRUE)
+  }
+
+  # Add lines to file
+  append_lines <- function(path, lines) {
+    con <- file(path, open = "a", encoding = "UTF-8")
+    on.exit(close(con), add = TRUE)
+    writeLines(lines, con = con, sep = "\n", useBytes = TRUE)
+  }
+
+  # Setting paths
+  if (!dir.exists(out_path)) dir.create(out_path, recursive = TRUE, showWarnings = FALSE)
+  metadata_parquet <- normalizePath(metadata_parquet)
+  metadata <- arrow::read_parquet(metadata_parquet)
+  md_path <- file.path(out_path, "amr_metadata_summary.md")
+
+  # Validation (got any data?)
+  if (nrow(metadata) == 0) {
+    stop("The output table is empty. Please check your query or input data.")
+  }
+
+  # get Species name
+  Species_name <- metadata |>
+    dplyr::distinct(genome.species) |>
+    dplyr::pull()
+
+  # Core summaries
+  TotalEntryCount <- metadata |> dplyr::count()
+  CleanEntryCount <- metadata |>
+    dplyr::distinct(genome.genome_id) |>
+    dplyr::count()
+
+  Antibiotics <- clean_distinct(metadata, genome_drug.antibiotic)
+  AntibioticClasses <- clean_distinct(metadata, drug_class)
+
+  LabMethods <- metadata |>
+    dplyr::mutate(
+      genome_drug.laboratory_typing_method = dplyr::case_when(
+        is.na(genome_drug.laboratory_typing_method) ~ "Not defined",
+        genome_drug.laboratory_typing_method == "" ~ "Not defined",
+        TRUE ~ genome_drug.laboratory_typing_method
+      )
+    ) |>
+    dplyr::count(genome_drug.laboratory_typing_method)
+
+  PubMed_ids <- clean_distinct(metadata, genome_drug.pmid)
+
+  PhenotypeCount <- metadata |>
+    dplyr::group_by(genome_drug.resistant_phenotype) |>
+    dplyr::count() |>
+    dplyr::arrange(-n) |>
+    dplyr::ungroup()
+
+  ## Stats by drugs
+  PhenotypebyDrugCount <- metadata |>
+    dplyr::group_by(genome_drug.resistant_phenotype, genome_drug.antibiotic) |>
+    dplyr::count() |>
+    dplyr::arrange(-n) |>
+    dplyr::ungroup()
+
+  ResPropbyDrug <- metadata |>
+    dplyr::group_by(genome_drug.antibiotic) |>
+    dplyr::count(genome_drug.resistant_phenotype) |>
+    dplyr::mutate(prop = n / sum(n)) |>
+    dplyr::filter(genome_drug.resistant_phenotype == "Resistant") |>
+    dplyr::transmute(genome_drug.antibiotic, res_prop = round(prop, 3)) |>
+    dplyr::arrange(-res_prop) |>
+    dplyr::ungroup()
+
+ # Collapse to one phenotype per genome x drug_class
+drug_class_calls <- metadata |>
+  dplyr::group_by(genome.genome_id, drug_class) |>
+  dplyr::summarise(
+    genome_drug.resistant_phenotype = dplyr::case_when(
+      any(genome_drug.resistant_phenotype == "Resistant", na.rm = TRUE) ~ "Resistant",
+      any(genome_drug.resistant_phenotype == "Intermediate", na.rm = TRUE) ~ "Intermediate",
+      any(genome_drug.resistant_phenotype == "Susceptible", na.rm = TRUE) ~ "Susceptible",
+      TRUE ~ NA_character_
+    ),
+    .groups = "drop"
+  )
+
+# Stats by drug class
+PhenotypebyDrugClassCount <- drug_class_calls |>
+  dplyr::count(genome_drug.resistant_phenotype, drug_class) |>
+  dplyr::arrange(dplyr::desc(n))
+
+ResPropbyDrugClass <- drug_class_calls |>
+  dplyr::group_by(drug_class) |>
+  dplyr::count(genome_drug.resistant_phenotype) |>
+  dplyr::mutate(prop = n / sum(n)) |>
+  dplyr::filter(genome_drug.resistant_phenotype == "Resistant") |>
+  dplyr::transmute(drug_class, res_prop = round(prop, 3)) |>
+  dplyr::arrange(dplyr::desc(res_prop)) |>
+  dplyr::ungroup()
+  
+  ## Collection years
+  Year <- metadata |>
+    dplyr::distinct(genome.collection_year) |>
+    dplyr::filter(!is.na(genome.collection_year)) |>
+    dplyr::pull() |>
+    sort()
+
+  YearCount <- metadata |>
+    dplyr::group_by(genome.collection_year) |>
+    dplyr::filter(!is.na(genome.collection_year)) |>
+    dplyr::count() |>
+    dplyr::arrange(-n) |>
+    dplyr::ungroup()
+
+  ## Geographical regions (countries)
+  Country <- clean_distinct(metadata, genome.isolation_country)
+  CountryCount <- metadata |>
+    dplyr::group_by(genome.isolation_country) |>
+    dplyr::filter(!is.na(genome.isolation_country), genome.isolation_country != "") |>
+    dplyr::count() |>
+    dplyr::arrange(-n) |>
+    dplyr::ungroup()
+
+  ## Isolation sources
+  Source <- clean_distinct(metadata, genome.isolation_source)
+  SourceCount <- metadata |>
+    dplyr::group_by(genome.isolation_source) |>
+    dplyr::filter(!is.na(genome.isolation_source), genome.isolation_source != "") |>
+    dplyr::count() |>
+    dplyr::arrange(-n) |>
+    dplyr::ungroup()
+
+  Host <- clean_distinct(metadata, genome.host_common_name)
+  HostCount <- metadata |>
+    dplyr::group_by(genome.host_common_name) |>
+    dplyr::filter(!is.na(genome.host_common_name), genome.host_common_name != "") |>
+    dplyr::count() |>
+    dplyr::arrange(-n) |>
+    dplyr::ungroup()
+
+  # Header
+  write_new(
+    md_path,
+    sprintf("# AMR summary report for *%s*", Species_name)
+  )
+
+  # Basic stats
+  append_lines(
+    md_path,
+    c(
+      sprintf("- **Total no. of observations**: %s", TotalEntryCount[[1]]),
+      sprintf("- **Unique genome IDs**: %s", CleanEntryCount[[1]]),
+      "",
+      sprintf(
+        "- **Associated publications** (%d): %s",
+        length(PubMed_ids),
+        if (length(PubMed_ids)) paste(PubMed_ids, collapse = ", ") else "None"
+      ),
+      ""
+    )
+  )
+
+  # Lists
+  append_lines(
+    md_path,
+    c(
+      sprintf("## Antibiotics (%d)", length(Antibiotics)),
+      "",
+      paste(Antibiotics, collapse = ", "),
+      "",
+      sprintf("## Antibiotic classes (%d)", length(AntibioticClasses)),
+      "",
+      paste(AntibioticClasses, collapse = ", "),
+      ""
+    )
+  )
+
+  # Tables!
+  append_lines(md_path, c("## Phenotype counts", "", md_tbl(PhenotypeCount), "", ""))
+  append_lines(md_path, c("## Phenotypes x antibiotic(s)", "", md_tbl(PhenotypebyDrugCount), "", ""))
+  append_lines(md_path, c("## Resistant proportions per antibiotic", "", md_tbl(ResPropbyDrug), "", ""))
+  append_lines(md_path, c("## Phenotypes x antibiotic class(es)", "", md_tbl(PhenotypebyDrugClassCount), "", ""))
+  append_lines(md_path, c("## Resistant proportion per antibiotic class", "", md_tbl(ResPropbyDrugClass), "", ""))
+  append_lines(md_path, c("## Laboratory methods", "", md_tbl(LabMethods), "", ""))
+  append_lines(md_path, c("## Collection years", "", md_tbl(YearCount), "", ""))
+  append_lines(md_path, c("## Isolation countries", "", md_tbl(CountryCount), "", ""))
+  append_lines(md_path, c("## Isolation sources", "", md_tbl(SourceCount), "", ""))
+  append_lines(md_path, c("## Hosts", "", md_tbl(HostCount), "", ""))
+
+
+  # Hosts as a simple list
+  # if (length(Host)) {
+  #   append_lines(md_path, c("## Hosts", "", paste0("- ", Host), "", ""))
+  # }
+}
+
+
+#' Write all summary plots to file(s)
+#'
+#' Expects `metadata_parquet` to be the output of `runDataProcessing()`'s
+#' `cleanData()` step (or an export of the resulting `metadata` table), since
+#' `drug_abbr`, `drug_class`, and `num_resistant_classes` are only populated
+#' after that step joins in the reference drug tables.
+#'
+#' @param metadata_parquet Character. Path to the Parquet metadata file.
+#' @param out_path Character. Output directory for plot files.
+#'
+#' @return Invisibly returns the path to the written PDF (all plots as
+#'   separate pages of one multi-page file).
+#' @export
+generatePlots <- function(metadata_parquet,
+                          out_path) {
+  if (!dir.exists(out_path)) {
+    dir.create(out_path, showWarnings = FALSE, recursive = TRUE)
+  }
+
+  metadata <- arrow::read_parquet(normalizePath(metadata_parquet))
+
+  if (nrow(metadata) == 0) {
+    stop("The input metadata table is empty. Please check your query or input data.")
+  }
+
+  ## Generate plots
+  # 1) Phenotypes across antibiotics and time
+  df_year <- metadata |>
+    dplyr::filter(!is.na(genome.collection_year)) |>
+    dplyr::select(
+      genome.genome_id,
+      drug_abbr,
+      genome_drug.resistant_phenotype,
+      genome.isolation_country,
+      genome.collection_year
+    )
+
+  summary_year <- df_year |>
+    dplyr::group_by(
+      drug_abbr,
+      genome_drug.resistant_phenotype,
+      genome.collection_year
+    ) |>
+    dplyr::summarise(count = dplyr::n(), .groups = "drop")
+
+  p1 <- ggplot2::ggplot(
+    summary_year,
+    ggplot2::aes(
+      x = genome.collection_year,
+      y = count,
+      colour = genome_drug.resistant_phenotype
+    )
+  ) +
+    ggplot2::geom_line() +
+    ggplot2::geom_point() +
+    ggplot2::facet_wrap(~drug_abbr, scales = "free_y") +
+    ggplot2::labs(
+      title = "Distribution of AMR phenotypes across antibiotics by year",
+      x = "Year", y = "Number of isolates",
+      colour = "Phenotype"
+    ) +
+    ggplot2::scale_color_manual(values = PHENOTYPE_COLORS, na.value = "gray70") +
+    ggplot2::theme_minimal(base_size = 12) +
+    ggplot2::theme(
+      text = ggplot2::element_text(colour = "black"),
+      legend.position = "bottom",
+      axis.text.x = ggplot2::element_text(angle = 45, hjust = 1, colour = "black"),
+      axis.text.y = ggplot2::element_text(colour = "black"),
+      axis.title = ggplot2::element_text(colour = "black"),
+      panel.grid.minor = ggplot2::element_blank()
+    )
+
+  # 2) Resistance only over time, colored by antibiotic (named palette aligned
+  # to factor levels via meta_palette() so colors stay consistent across runs)
+  abx_levels <- summary_year |>
+    dplyr::filter(genome_drug.resistant_phenotype == "Resistant") |>
+    dplyr::distinct(drug_abbr) |>
+    dplyr::arrange(drug_abbr) |>
+    dplyr::pull(drug_abbr)
+
+  pal_named <- stats::setNames(meta_palette(length(abx_levels)), abx_levels)
+
+  p2 <- ggplot2::ggplot(
+    summary_year |>
+      dplyr::filter(genome_drug.resistant_phenotype == "Resistant") |>
+      dplyr::mutate(
+        antibiotic_fac = factor(drug_abbr, levels = abx_levels)
+      ),
+    ggplot2::aes(
+      x = genome.collection_year, y = count,
+      colour = antibiotic_fac
+    )
+  ) +
+    ggplot2::geom_line() +
+    ggplot2::geom_point() +
+    ggplot2::labs(
+      title  = "Distribution of resistant isolates by year",
+      x      = "Year",
+      y      = "Number of AMR isolates",
+      colour = "Antibiotic"
+    ) +
+    ggplot2::scale_color_manual(values = pal_named, drop = TRUE) + # <- named palette
+    ggplot2::theme_minimal(base_size = 12) +
+    ggplot2::theme(
+      text = ggplot2::element_text(colour = "#2D2D2D"),
+      legend.position = "bottom"
+    )
+
+  # 3) Time × geography × phenotype
+  df_country <- metadata |>
+    dplyr::filter(genome.isolation_country != "") |>
+    dplyr::select(
+      genome.genome_id,
+      drug_abbr,
+      genome_drug.resistant_phenotype,
+      genome.isolation_country,
+      genome.collection_year
+    )
+
+  summary_country_year <- df_country |>
+    dplyr::group_by(
+      genome.collection_year,
+      genome_drug.resistant_phenotype,
+      genome.isolation_country
+    ) |>
+    dplyr::summarise(count = dplyr::n(), .groups = "drop")
+
+  p3 <- ggplot2::ggplot(
+    summary_country_year,
+    ggplot2::aes(
+      x = genome.collection_year,
+      y = genome.isolation_country,
+      size = count,
+      color = genome_drug.resistant_phenotype
+    )
+  ) +
+    ggplot2::geom_point(alpha = 0.75) +
+    ggplot2::scale_size(range = c(3, 15)) +
+    ggplot2::scale_color_manual(values = PHENOTYPE_COLORS, na.value = "gray70") +
+    ggplot2::labs(
+      title = "Distribution of AMR phenotype by year and geography",
+      x = "Year", y = "Country",
+      size = "Count", color = "Phenotype"
+    ) +
+    ggplot2::theme_minimal(base_size = 12) +
+    ggplot2::theme(
+      text = ggplot2::element_text(colour = "black"),
+      legend.position = "right"
+    )
+
+  # 4) Phenotype proportion per antibiotic (stacked, normalized)
+  p4 <- ggplot2::ggplot(
+    metadata,
+    ggplot2::aes(
+      x = drug_abbr,
+      fill = genome_drug.resistant_phenotype
+    )
+  ) +
+    ggplot2::geom_bar(position = "fill") +
+    ggplot2::coord_flip() +
+    ggplot2::labs(
+      title = "Distribution of AMR phenotypes",
+      x = "Antibiotic", y = "Proportion", fill = "Phenotype"
+    ) +
+    ggplot2::scale_fill_manual(values = PHENOTYPE_COLORS, na.value = "gray70") +
+    ggplot2::theme_minimal(base_size = 12) +
+    ggplot2::theme(
+      text = ggplot2::element_text(colour = "black"),
+      legend.position = "bottom",
+      axis.text.x = ggplot2::element_text(angle = 45, hjust = 1)
+    )
+
+  # 5) Resistant isolates by isolation source (top 10 sources + Other, same
+  # convention as amRviz's makeIsolationSourcesPlot)
+  summary_isolation_source <- metadata |>
+    dplyr::filter(genome.isolation_source != "") |>
+    dplyr::group_by(genome.isolation_source, genome_drug.resistant_phenotype) |>
+    dplyr::summarise(count = dplyr::n(), .groups = "drop") |>
+    dplyr::filter(genome_drug.resistant_phenotype == "Resistant")
+
+  top_isolation_sources <- summary_isolation_source |>
+    dplyr::slice_max(order_by = count, n = 10) |>
+    dplyr::pull(genome.isolation_source)
+
+  summary_isolation_source <- summary_isolation_source |>
+    dplyr::mutate(
+      isolation_source_grp = ifelse(
+        genome.isolation_source %in% top_isolation_sources,
+        genome.isolation_source,
+        "Other"
+      )
+    )
+
+  n_sources <- dplyr::n_distinct(summary_isolation_source$isolation_source_grp)
+
+  p5 <- ggplot2::ggplot(
+    summary_isolation_source,
+    ggplot2::aes(
+      x = stats::reorder(isolation_source_grp, count),
+      y = count, fill = isolation_source_grp
+    )
+  ) +
+    ggplot2::geom_col() +
+    ggplot2::coord_flip() +
+    ggplot2::labs(
+      title = "Distribution of AMR isolates by isolation source",
+      x = "Isolation source", y = "Number of AMR isolates"
+    ) +
+    ggplot2::scale_fill_manual(values = meta_palette(n_sources)) +
+    ggplot2::theme_minimal(base_size = 12) +
+    ggplot2::theme(
+      text = ggplot2::element_text(colour = "black"),
+      legend.position = "none"
+    )
+
+  # 6) Histogram of resistant classes per genome
+  p6 <- ggplot2::ggplot(metadata, ggplot2::aes(num_resistant_classes)) +
+    ggplot2::geom_histogram(binwidth = 1, fill = META_COLORS[[1]]) +
+    ggplot2::labs(
+      title = "Distribution of AMR classes per isolate",
+      x = "# Resistant Classes", y = "Count"
+    ) +
+    ggplot2::theme_minimal(base_size = 12) +
+    ggplot2::theme(
+      text = ggplot2::element_text(colour = "black"),
+      legend.position = "bottom",
+      axis.text = ggplot2::element_text(colour = "black"),
+      axis.title = ggplot2::element_text(colour = "black"),
+      panel.grid.minor = ggplot2::element_blank()
+    )
+
+  plots <- list(p1 = p1, p2 = p2, p3 = p3, p4 = p4, p5 = p5, p6 = p6)
+
+  # Table for drug name - abbreviation mapping
+  drug_table <- metadata |>
+    dplyr::distinct(genome_drug.antibiotic, drug_abbr) |>
+    dplyr::rename(
+      "Abbreviation" = "drug_abbr",
+      "Antibiotic" = "genome_drug.antibiotic"
+    ) |>
+    dplyr::arrange(Antibiotic)
+
+  ## Write to device
+  pdf_path <- file.path(out_path, "amRdata_exploratory_plots.pdf")
+  grDevices::pdf(pdf_path, onefile = TRUE)
+  on.exit(grDevices::dev.off(), add = TRUE)
+  for (nm in names(plots)) {
+    print(plots[[nm]])
+  }
+  # grid::grid.newpage()
+
+  gridExtra::grid.arrange(
+    grid::textGrob(
+      "Antibiotic name abbreviations",
+      gp = grid::gpar(fontsize = 16, fontface = "bold")
+    ),
+    gridExtra::tableGrob(drug_table, rows = NULL),
+    ncol = 1,
+    heights = c(0.08, 0.92)
+  )
+
+  invisible(pdf_path)
+}
diff --git a/R/utils_colors.R b/R/utils_colors.R
new file mode 100644
index 0000000..7c4a25e
--- /dev/null
+++ b/R/utils_colors.R
@@ -0,0 +1,41 @@
+# Colour palettes shared across the amR suite's plots (kept in sync with
+# amRviz's R/utils_colors.R so reports and the dashboard read as one system).
+
+# AMR phenotype palette (R/S/I plus full-word + lowercase variants so it works
+# regardless of how the column is encoded). Susceptible is intentionally
+# neutral grey so Resistant amber stands out as the signal of interest.
+PHENOTYPE_COLORS <- c(
+  R = "#d4872a", Resistant = "#d4872a", resistant = "#d4872a",
+  S = "#5b8db8", Susceptible = "#5b8db8", susceptible = "#5b8db8",
+  I = "#e6ab80", Intermediate = "#e6ab80", intermediate = "#e6ab80"
+)
+
+# Categorical palette for metadata plots (hosts, isolation sources, etc.).
+# Use `meta_palette(n)` at call sites rather than referencing META_COLORS
+# directly, so plots handle more categories than length(META_COLORS) without
+# falling back to NA / gray70.
+META_COLORS <- c(
+  "#4a6b8a", "#87ceeb", "#e6ab80", "#8b6b7a",
+  "#4e9a9a", "#c4a35a", "#9b7fba", "#5b8db8",
+  "#d4735e", "#d4872a", "#b5b5b5"
+)
+
+#' Categorical palette of `n` muted colors
+#'
+#' For `n <= length(META_COLORS)` returns the first `n` curated colors verbatim;
+#' above that, interpolates via `grDevices::colorRampPalette()` so callers get a
+#' full palette regardless of how many unique values their data has.
+#'
+#' @param n Number of colors to return.
+#' @return A character vector of `n` hex colors (empty when `n <= 0`).
+#' @keywords internal
+#' @noRd
+meta_palette <- function(n = length(META_COLORS)) {
+  if (n <= 0) {
+    return(character(0))
+  }
+  if (n <= length(META_COLORS)) {
+    return(META_COLORS[seq_len(n)])
+  }
+  grDevices::colorRampPalette(META_COLORS)(n)
+}
diff --git a/data/amRdata_exploratory_plots.pdf b/data/amRdata_exploratory_plots.pdf
new file mode 100644
index 0000000..f54769b
Binary files /dev/null and b/data/amRdata_exploratory_plots.pdf differ
diff --git a/data/amr_metadata_summary.md b/data/amr_metadata_summary.md
new file mode 100644
index 0000000..b280f57
--- /dev/null
+++ b/data/amr_metadata_summary.md
@@ -0,0 +1,150 @@
+# AMR summary report for *Staphylococcus argenteus*
+- **Total no. of observations**: 723
+- **Unique genome IDs**: 73
+
+- **Associated publications** (2): 26806258, 34100643
+
+## Antibiotics (16)
+
+cefoxitin, chloramphenicol, ciprofloxacin, clindamycin, daptomycin, erythromycin, fosfomycin, fusidic_acid, gentamicin, levofloxacin, linezolid, oxacillin, penicillin, tetracycline, trimethoprim-sulfamethoxazole, vancomycin
+
+## Antibiotic classes (14)
+
+aminoglycosides, amphenicols, cephalosporins, fluoroquinolones, fusidane, glycopeptides, lincosamides, lipopeptides, macrolides, oxazolidinones, penicillins, phosphonics, tetracyclines, trimethoprim-sulfonamides
+
+## Phenotype counts
+
+|genome_drug.resistant_phenotype |   n|
+|:-------------------------------|---:|
+|Susceptible                     | 633|
+|Resistant                       |  90|
+
+
+## Phenotypes x antibiotic(s)
+
+|genome_drug.resistant_phenotype |genome_drug.antibiotic        |  n|
+|:-------------------------------|:-----------------------------|--:|
+|Susceptible                     |gentamicin                    | 73|
+|Susceptible                     |cefoxitin                     | 72|
+|Susceptible                     |oxacillin                     | 72|
+|Susceptible                     |trimethoprim-sulfamethoxazole | 72|
+|Resistant                       |penicillin                    | 69|
+|Susceptible                     |clindamycin                   | 68|
+|Susceptible                     |fusidic_acid                  | 68|
+|Susceptible                     |erythromycin                  | 67|
+|Susceptible                     |fosfomycin                    | 58|
+|Susceptible                     |levofloxacin                  | 58|
+|Resistant                       |erythromycin                  |  5|
+|Susceptible                     |daptomycin                    |  5|
+|Susceptible                     |linezolid                     |  5|
+|Susceptible                     |vancomycin                    |  5|
+|Resistant                       |clindamycin                   |  4|
+|Susceptible                     |penicillin                    |  4|
+|Resistant                       |chloramphenicol               |  3|
+|Resistant                       |ciprofloxacin                 |  3|
+|Resistant                       |tetracycline                  |  3|
+|Susceptible                     |chloramphenicol               |  2|
+|Susceptible                     |ciprofloxacin                 |  2|
+|Susceptible                     |tetracycline                  |  2|
+|Resistant                       |cefoxitin                     |  1|
+|Resistant                       |oxacillin                     |  1|
+|Resistant                       |trimethoprim-sulfamethoxazole |  1|
+
+
+## Resistant proportions per antibiotic
+
+|genome_drug.antibiotic        | res_prop|
+|:-----------------------------|--------:|
+|penicillin                    |    0.945|
+|chloramphenicol               |    0.600|
+|ciprofloxacin                 |    0.600|
+|tetracycline                  |    0.600|
+|erythromycin                  |    0.069|
+|clindamycin                   |    0.056|
+|cefoxitin                     |    0.014|
+|oxacillin                     |    0.014|
+|trimethoprim-sulfamethoxazole |    0.014|
+
+
+## Phenotypes x antibiotic class(es)
+
+|genome_drug.resistant_phenotype |drug_class                |  n|
+|:-------------------------------|:-------------------------|--:|
+|Susceptible                     |aminoglycosides           | 73|
+|Susceptible                     |cephalosporins            | 72|
+|Susceptible                     |trimethoprim-sulfonamides | 72|
+|Resistant                       |penicillins               | 69|
+|Susceptible                     |fusidane                  | 68|
+|Susceptible                     |lincosamides              | 68|
+|Susceptible                     |macrolides                | 67|
+|Susceptible                     |fluoroquinolones          | 60|
+|Susceptible                     |phosphonics               | 58|
+|Resistant                       |macrolides                |  5|
+|Susceptible                     |glycopeptides             |  5|
+|Susceptible                     |lipopeptides              |  5|
+|Susceptible                     |oxazolidinones            |  5|
+|Resistant                       |lincosamides              |  4|
+|Susceptible                     |penicillins               |  4|
+|Resistant                       |amphenicols               |  3|
+|Resistant                       |fluoroquinolones          |  3|
+|Resistant                       |tetracyclines             |  3|
+|Susceptible                     |amphenicols               |  2|
+|Susceptible                     |tetracyclines             |  2|
+|Resistant                       |cephalosporins            |  1|
+|Resistant                       |trimethoprim-sulfonamides |  1|
+
+
+## Resistant proportion per antibiotic class
+
+|drug_class                | res_prop|
+|:-------------------------|--------:|
+|penicillins               |    0.945|
+|amphenicols               |    0.600|
+|tetracyclines             |    0.600|
+|macrolides                |    0.069|
+|lincosamides              |    0.056|
+|fluoroquinolones          |    0.048|
+|cephalosporins            |    0.014|
+|trimethoprim-sulfonamides |    0.014|
+
+
+## Laboratory methods
+
+|genome_drug.laboratory_typing_method         |   n|
+|:--------------------------------------------|---:|
+|Biofosun Gram-positive panels broth dilution |  65|
+|Not defined                                  | 658|
+
+
+## Collection years
+
+| genome.collection_year|   n|
+|----------------------:|---:|
+|                   2015| 671|
+|                   2017|  39|
+|                   2018|  13|
+
+
+## Isolation countries
+
+|genome.isolation_country |   n|
+|:------------------------|---:|
+|Thailand                 | 658|
+
+
+## Isolation sources
+
+|genome.isolation_source |  n|
+|:-----------------------|--:|
+|pork                    | 39|
+|duck                    | 26|
+
+
+## Hosts
+
+|genome.host_common_name |   n|
+|:-----------------------|---:|
+|Human                   | 658|
+|Duck                    |  26|
+
+
diff --git a/data/metadata.parquet b/data/metadata.parquet
new file mode 100644
index 0000000..ff600a0
Binary files /dev/null and b/data/metadata.parquet differ
diff --git a/man/generatePlots.Rd b/man/generatePlots.Rd
new file mode 100644
index 0000000..82845a2
--- /dev/null
+++ b/man/generatePlots.Rd
@@ -0,0 +1,23 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/amRdataPlots.R
+\name{generatePlots}
+\alias{generatePlots}
+\title{Write all summary plots to file(s)}
+\usage{
+generatePlots(metadata_parquet, out_path)
+}
+\arguments{
+\item{metadata_parquet}{Character. Path to the Parquet metadata file.}
+
+\item{out_path}{Character. Output directory for plot files.}
+}
+\value{
+Invisibly returns the path to the written PDF (all plots as
+separate pages of one multi-page file).
+}
+\description{
+Expects \code{metadata_parquet} to be the output of \code{runDataProcessing()}'s
+\code{cleanData()} step (or an export of the resulting \code{metadata} table), since
+\code{drug_abbr}, \code{drug_class}, and \code{num_resistant_classes} are only populated
+after that step joins in the reference drug tables.
+}
diff --git a/man/generateSummary.Rd b/man/generateSummary.Rd
new file mode 100644
index 0000000..bbe49ef
--- /dev/null
+++ b/man/generateSummary.Rd
@@ -0,0 +1,28 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/amRdataPlots.R
+\name{generateSummary}
+\alias{generateSummary}
+\title{Generate a summary report for AMR metadata}
+\usage{
+generateSummary(metadata_parquet, out_path)
+}
+\arguments{
+\item{metadata_parquet}{Character string. Path to a Parquet file containing
+standardized AMR metadata.}
+
+\item{out_path}{Character string. Directory where the Markdown report is written.}
+}
+\value{
+Writes a structured, human‑readable summary report to
+"<out_path>/amr_metadata_summary.md".
+}
+\description{
+Generate a summary report for AMR metadata
+}
+\examples{
+generateSummary(
+  metadata_parquet = "results/metadata.parquet",
+  out_path = "results/"
+)
+
+}