A workflow to benchmark tools for differential translation efficiency analysis (TEA) and a companion repository to
Rp-Bp and the Ribotools box for translatome profiling, Etienne Boileau, Maja Bencun, Francesca Tuorto, Isabel S. Naarmann-de Vries, Pauline Fahjen, Philipp Mertins, and Christoph Dieterich.
This repository is a supplementary companion, not a standalone package. It's key use case is to benchmark selected TEA software tools using the simulated data from Oertlin et al. and a subset of the data from Boileau et al..
Until recently, this was a well-established field with numerous published benchmarks. However, as of 2026, many existing tools have been deprecated and are now obsolete or no longer actively maintained. A new generation of end-to-end (E2E) software, such as Ribotools, is making transcriptomics analysis more accessible, reproducible, and interpretable. For further details, results, and discussion, refer to the manuscript.
# install snakemake into an environment named "benchmark"
mamba create -c conda-forge -c bioconda -c nodefaults -n benchmark snakemake-minimal
# install conda into this environment
mamba activate benchmark
mamba install -c conda-forge condaNote: Setting conda config --set channel_priority strict may break environment specs for selected tools. Use the default settings or
explicitly set channel priority to flexible.
To configure this workflow, modify the following files:
- config/config.yaml: workflow-specific parameters
This use case is self-documenting. If you want to add a new tool to the benchmark, update the configuration file, add an environment file e.g. workflow/envs/ribotools_lrt.yaml and a corresponding script e.g. workflow/scripts/run_ribotools_lrt.
While benchmark rules can handle any number of tools, each tool must be called by a dedicated R script with strict I/O formatting guidelines. Each benchmark is associated with specific input data and parameters, which must match rule descriptions and evaluation scripts. More complex use cases may require additional modifications. For more hints, consult docs/DESCRIPTION.md.
Community contributions are welcome.
Snakemake scheduling hints are set in a default profile under workflow/profiles/default/config.yaml. In addition, processes memory and CPU usage (e.g. BLAS and OMP used via parallel/BiocParallel by some R packages) are capped at the OS level in Snakemake rules.
snakemake --profile workflow/profiles/default