Analyse many plates — analyze

Batch wrapper around analyze_plate() that never stops on a single bad image: failures are recorded and reported. A QC figure is written for every plate — inspect them before using the numbers.

Usage

analyze_plates(
  files,
  metadata = NULL,
  layout = plate_layout(),
  qc_dir = NULL,
  ...,
  verbose = TRUE,
  coefficient_control = "control",
  coefficient_by = NULL
)

Arguments

files: Character vector of image paths, or a directory (all JPEG/PNG/TIFF files are used).
metadata: Optional design data frame with one row per image. Image identities are validated; missing plate IDs are generated. Treatment, set and context fields may remain blank for measurement. Matched coefficients are withheld until a valid experimental design is available. Extra design columns are merged into the output tables.
layout: A single plate_layout(), or a function function(metadata_row) returning the layout for each plate (e.g. centre = "none" for controls).
qc_dir: Directory for QC PNGs (created if needed). The default NULL writes nothing; give a folder, e.g. qc_dir = "qc", to save one QC figure per plate.
...: Further arguments passed to analyze_plate().
verbose: Print progress.
coefficient_control: Control treatment label for automatic coefficient calculation; NULL disables it. Calculated when treatment metadata exists.
coefficient_by: Matching columns for controls, e.g. strain and day.

Value

A list of three data frames (colonies, profiles, plates) and a failed data frame. Also contains melanization_coefficient (coefficient table), coefficient_plate_means and coefficient_error. Coefficient errors preserve image results but must be resolved before use.

Examples

if (FALSE) { # \dontrun{
meta <- data.frame(file = c("2026_05_26_20.JPG", "2026_05_26_37.JPG"),
                   plate_id = c("t1", "c1"), set_id = "set1",
                   treatment = c("fungus_A", "control"))
lay <- function(row) plate_layout(centre = if (row$treatment == "control") "none" else "fungus")
out <- analyze_plates("plates/", metadata = meta, layout = lay, qc_dir = "qc/")
write.csv(out$colonies, "colonies.csv", row.names = FALSE)
} # }