Install OncoTracer¶
Use 64-bit Linux and Python 3.10–3.13. Install the application, then choose Conda or Docker for analysis tools.
Quick start¶
pip install oncotracer
No Git checkout is needed. If system Python requires a separate environment, use the Python environment instructions below.
Conda¶
Install Conda/Miniforge first, then:
oncotracer install --conda
oncotracer doctor --backend conda
oncotracer --interactive --backend conda
Installation prepares alignment, QDNAseq, ichorCNA, classifier/reporting, GISTIC and verified Illumina variant environments. Setup selects their tools automatically. FFPErase, Strelka2 and ONT resources need additional setup.
Docker¶
Docker must be running and accessible to your user. Host Conda is unnecessary:
oncotracer install --docker --image carlosfarkas/oncotracer:terminal-progress-20261022
oncotracer doctor --backend docker
oncotracer --interactive --backend docker
Installation saves the selected image; doctor and setup reuse it. Select small variants in the browser when needed. See Docker variants for models, FFPERASE licensing and ANNOVAR. Use Conda for ONT methylation.
Start a project¶
Use the complete printed browser URL and keep its terminal open. Choose Illumina or ONT, then Fresh or FFPE. Select inputs, assign samples to Normal or Cancer, choose a project folder and review settings. Click Save configuration and check, then Run analysis.
For a named project:
oncotracer --interactive --backend conda --project /absolute/path/to/my-study
Use --backend docker for Docker. Separate terminals can run separate projects at the same time. Each session gets its own URL and available port; give each run a distinct project/output folder and a share of CPU and memory.
For terminal prompts, use oncotracer setup --terminal --backend conda. To reopen a saved project, use oncotracer setup --project /absolute/path/to/my-study --run. To check and run its saved configuration separately:
oncotracer check --config /absolute/path/to/my-study/config/run.yml
oncotracer run --backend conda --config /absolute/path/to/my-study/config/run.yml
Runs use prebuilt hg38 indexes by default. See genome indexes, the setup guide, or terminal and SSH instructions.
Nextflow entry¶
If you have Nextflow and an OncoTracer source checkout, supply its full path from any directory:
nextflow run /absolute/path/to/oncotracer-src/main.nf --interactive --backend conda
Use --backend docker for Docker. See Nextflow setup.
Python environment¶
For a dedicated application environment:
python3 -m venv oncotracer-env
source oncotracer-env/bin/activate
python -m pip install oncotracer
Keep this environment. Backend installation creates an extra ~/.local/bin/oncotracer command when needed and preserves a pip-owned command already there. Follow the installer’s PATH instruction if necessary; --no-launcher opts out. See advanced installation for source, standalone, Poetry and Apptainer routes.
Requirements¶
For small runs with 2–4 threads, plan for 16 GiB available RAM for Illumina copy-number analysis or 24 GiB for ONT. Models and larger runs need more. Leave at least 60 GiB free for tools, references and a small run, plus input reads, BAMs and temporary files; Docker extraction needs additional space. Check your intended storage:
oncotracer system --path /absolute/path/to/my-study
Desktop file browsing uses Zenity or KDialog; paths can also be entered manually. Remote sessions use server paths.
Uninstall¶
Permanently remove managed Conda tools:
oncotracer uninstall --conda --yes --purge
Then remove the application with python -m pip uninstall oncotracer in its Python environment. Docker images and any remaining extra global launcher have separate removal steps. See uninstall for the component commands and retained settings; projects and results remain.