
GalaxyR is an R package for programmatic interaction
with the Galaxy API (tested primarily against Galaxy Europe).
It allows you to manage histories, upload data, run tools and workflows,
wait for jobs to complete, and download results β all directly from
R.
This package is designed for automation, reproducibility, and scripting, not UI replacement.
Install this package from CRAN:
install.packages("GalaxyR")Or install the latest version directly from GitHub:
# install.packages("remotes")
remotes::install_github("JulFrey/GalaxyR")Before using the package, you must set your Galaxy API key.
You can either:
galaxy_set_credentials("your-secret-key", galaxy_url = "https://usegalaxy.eu")~/.Renviron (recommended)#usethis::edit_r_environ()
GALAXY_API_KEY = "your-secret-key"
GALAXY_URL = "https://usegalaxy.eu"Restart R after editing.Renviron.
The default Galaxy instance is:
"https://usegalaxy.eu"The Galaxy Insatance can be specified using the
galaxy_set_credentials() function or the
.Renviron file (see above). All functions also accept a
galaxy_url argument if you want to target a different
Galaxy server.
Below is a complete example that:
# Load GalaxyR
library(GalaxyR)
# Get the tool ID and inspect inputs
tool <- galaxy_get_tool_id("Add line to file")
inputs <- galaxy_get_tool(tool)
# Create a tiny test file
test_file <- tempfile(fileext = ".txt")
test_text <- "This is an example \ntest file."
writeLines(test_text,test_file)
# directory for outputs
outdir <- tempdir()
# Run on Galaxy
gxy <- galaxy(history_name = "add line example") |> # S4 class with history name
galaxy_initialize() |> # initialise Galaxy history
galaxy_upload_https(test_file) |> # upload test file
galaxy_run_tool(tool, inputs = list(text_input = "added example text")) |> # run
galaxy_poll_tool() |> # wait for completion
galaxy_download_result(outdir)
# Inspect the result
# Inspect the result
results <- list.files(outdir, full.names = TRUE)
readLines(results[grep("Add line to file", results)])Always use input name, not
label
Example: text_input, not
"text to add"
Dataset inputs must be passed as objects, not plain strings:
infile = list(
src = "hda",
id = DATASET_ID
)You can inspect expected inputs using:
galaxy_get_tool(tool_id)Galaxy jobs and datasets are asynchronous.
This package provides helpers to wait safely until execution
finishes: galaxy_poll_tool() waits for tool execution.
Terminal states: - β
ok - β error - ποΈ
deleted
| Function | Description |
|---|---|
galaxy_initialize() |
Create a new history |
galaxy_upload_https() |
Upload a file via HTTPS |
galaxy_run_tool() |
Run a Galaxy tool |
galaxy_poll_tool() |
Wait for tool completion |
galaxy_download_result() |
Download dataset |
galaxy_get_tool() |
Inspect tool metadata |
galaxy_list_tools() |
List installed tools |
galaxy_get_tool() |
Get all tool options and parameters |
galaxy_print_tool_inputs() |
Print a better overview of the inputs for a tool |
galaxy_history_size() |
Compute history disk usage |
galaxy_list_files() |
List all files within a history |
galaxy_delete_history() |
Delets or purges full histories for cleanup |
Author:
Julian Frey
Chair of Forest Growth and
Dendroecology
University of Freiburg