Running RAFT locally
Prerequisites
RAFT installs as a Python package, but running workflows locally also requires several external tools:
Python 3.9+
Nextflow
Singularity or Apptainer
Note
RAFT provides a --bypass-dependencies flag for advanced cases, but the
recommended setup is to install the required system dependencies normally.
Install RAFT
Install RAFT with pip:
$ pip install --user reproducible-analyses-framework-and-tools
Depending on your environment, you may need to add ~/.local/bin to your
$PATH.
Set up a workspace
Initialize the standard RAFT workspace in your current directory:
$ raft setup --default
This creates .raft.cfg and the standard directories such as:
projects/
references/
imgs/
shared/
inputs/
If you want to place those directories in custom locations, run:
$ raft setup
Try the demo
Run a demo project:
$ raft run-demo --workflow lens --version v1.9-dev
Run your own project
Create and run a project:
$ raft run \
--project-id my-project \
--workflow lens \
--version v1.9-dev \
--manifest my_manifest.tsv
Note
if --manifest is omitted, RAFT will launch the web-based manifest generator.
Note
if --workflow-profile is omitted, RAFT will launch the web-based workflow configuration tool.
If a project already exists, run it again with:
$ raft run --project-id my-project
Do not repeat setup-only flags such as --workflow, --version,
--manifest, or --setup-only when running an existing project. RAFT
treats those as project-creation options and will refuse to mix them with an
existing project run.
If you want workflow-specific recommended profiles in the documentation, those
should be listed on the workflow inventory page and passed via
--workflow-profile.
For additional setup and execution flags, see Helpful RAFT options. For common errors and next-step commands, see Troubleshooting.
Manage existing projects
After you have run one or more projects, use raft generate-reports to
review outputs and the project packaging/loading commands to transfer or archive
projects.