API reference

This page documents PETFit’s main R functions for launching apps and running pipelines.

Interactive apps

petfit_interactive()

Launch a PETFit Shiny app in your browser.

petfit_interactive(
  app = c("regiondef", "modelling_plasma", "modelling_ref"),
  bids_dir = NULL,
  derivatives_dir = NULL,
  blood_dir = NULL,
  petfit_output_foldername = "petfit",
  analysis_foldername = "Primary_Analysis",
  config_file = NULL,
  cores = 1L,
  save_logs = FALSE,
  ancillary_analysis_folder = NULL
)

Arguments:

Argument	Description
app	Which app to launch: "regiondef", "modelling_plasma", or "modelling_ref"
bids_dir	Path to the BIDS directory
derivatives_dir	Path to derivatives directory. Defaults to bids_dir/derivatives if bids_dir is provided
blood_dir	Path to blood data directory (for modelling_plasma)
petfit_output_foldername	Name of the petfit output folder within derivatives (default: "petfit")
analysis_foldername	Name of the analysis subfolder (default: "Primary_Analysis")
config_file	Path to an existing configuration file (for modelling apps)
cores	Number of cores for parallel processing (default: 1L)
save_logs	Whether to save processing logs (default: FALSE)
ancillary_analysis_folder	Name of a sibling analysis subfolder to inherit delay or k2prime estimates from. Must be a folder name (e.g. "Ancillary_Analysis"), not a full path

Automatic pipelines

petfit_auto()

Unified entry point for running any PETFit pipeline non-interactively.

petfit_auto(
  app = c("regiondef", "modelling_plasma", "modelling_ref"),
  bids_dir = NULL,
  derivatives_dir = NULL,
  blood_dir = NULL,
  petfit_output_foldername = "petfit",
  analysis_foldername = "Primary_Analysis",
  step = NULL,
  cores = 1L,
  save_logs = FALSE,
  ancillary_analysis_folder = NULL
)

Dispatches to petfit_regiondef_auto() or petfit_modelling_auto() based on app.

petfit_regiondef_auto()

Run region definition non-interactively.

petfit_regiondef_auto(
  bids_dir = NULL,
  derivatives_dir = NULL,
  petfit_output_foldername = "petfit",
  cores = 1L
)

petfit_modelling_auto()

Run the modelling pipeline non-interactively.

petfit_modelling_auto(
  bids_dir = NULL,
  derivatives_dir = NULL,
  petfit_output_foldername = "petfit",
  analysis_foldername = "Primary_Analysis",
  blood_dir = NULL,
  step = NULL,
  pipeline_type = NULL,
  cores = 1L,
  save_logs = FALSE,
  ancillary_analysis_folder = NULL
)

Argument	Description
step	Run a specific step: "datadef", "weights", "delay", "reference_tac", "model1", "model2", "model3". If NULL, runs all steps
pipeline_type	Explicit pipeline type: "plasma" or "reference". If NULL, auto-detected from the configuration file
ancillary_analysis_folder	Name of a sibling analysis subfolder to inherit delay or k2prime estimates from

All other arguments are the same as petfit_interactive().

Container command-line options

When running PETFit in Docker or Apptainer, the container accepts these flags:

Flag	Description
--func	App to run: regiondef, modelling_plasma, or modelling_ref (required)
--mode	interactive (default) or automatic
--step	Specific step for automatic mode (see petfit_modelling_auto() above)
--analysis_foldername	Analysis subfolder name (default: Primary_Analysis)
--petfit_output_foldername	Name of petfit output folder within derivatives (default: petfit)
--cores	Number of cores for parallel processing (default: 1)