RNAConSnake

RNAConSnake is a workflow for discovering and prioritizing locally conserved RNA secondary structure candidates from multiple-sequence alignments.

It combines RNALalifold candidate discovery with per-candidate filtering, conservation scoring, covariation analysis, structure prediction, and structured result export. RNAConSnake is intended as the scientific computation layer: it produces candidate alignments, figures, metrics, and summary tables, but it does not generate browser-style HTML reports.

RNAConSnake is released under the GNU Affero General Public License v3 or later (AGPL-3.0-or-later).

What Goes In

• one input multiple-sequence alignment in Stockholm format
• one or more maxbpspan window sizes
• an installed external RNA analysis toolchain available on PATH

An example input alignment is provided at examples/aln1.stk.

What Comes Out

RNAConSnake writes all outputs into a user-selected run directory. The main result classes are:

• split and cleaned candidate alignments in Stockholm and Clustal format
• per-candidate RNAalifold outputs and structure/alignment figures
• RNAz, alifoldz, max-covariation, and refold-derived metrics
• optional R-scape outputs
• per-candidate JSON summary records
• per-run summary files:
  • RNAConSnake.log
  • RNAConSnake.log.csv
  • RNAConSnake.md
• optional structured export bundles for downstream consumers

For a workflow-oriented overview, see docs/pipeline_summary.md.

Quickstart

Install in a virtual environment:

python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
pip install -e .[dev]

Check external dependencies:

rnaconsnake-run --check-deps

Run the workflow:

rnaconsnake-run --input-alignment /path/to/input.stk --output-dir /path/to/run_dir --cores all

If you want to use the repository checkout without activating the virtual environment, the repository-local launcher is:

./rnaconsnake-run --version

Documentation

• User guide: docs/usage.md
• Development notes: docs/development.md
• Pipeline summary: docs/pipeline_summary.md
• Architecture: ARCHITECTURE.md
• Contributing: CONTRIBUTING.md

Runtime Dependencies

RNAConSnake expects the external RNA analysis toolchain to be installed separately. The core workflow currently requires these commands on PATH:

• RNALalifold
• esl-reformat
• RNAz
• alifoldz.pl
• RNAalifold
• ps2eps
• epstopdf
• refold.pl
• RNAfold
• magick

Optional branches may additionally require tools such as R-scape, cmbuild, and cmcalibrate.

Development And Testing

The repository includes a pytest-based regression suite and a Snakemake workflow smoke test built around fake external tools. Run:

pytest -v

Detailed developer workflow notes are in docs/development.md.