Getting Started & Quick Reference

Launch your first siRNAforge workflow in minutes, then dive deeper through the focused docs sections.

1. Install

  • Python/uv users: Follow the step-by-step guide in Installation for pip/uv instructions plus development setup (make dev) that matches the project’s CI environment.

  • Container users: Pull or build the full bioinformatics image as described in Installation → Docker. The container bundles Nextflow, BWA-MEM2, SAMtools, ViennaRNA, and detects local mode automatically.

Need to automate lab servers or run air-gapped? The installation guide also documents offline caching, verification commands, and how to run smoke tests with the bundled FASTA files.

2. Run Your First Workflow

# End-to-end design + scoring for TP53
uv run sirnaforge workflow TP53 --output-dir tp53_run

  • Use docker run --rm ghcr.io/austin-s-h/sirnaforge:latest sirnaforge workflow … for the same behavior inside the prebuilt image.

  • The command performs gene search → siRNA design → thermodynamic scoring → filtering and writes CSVs plus logs under tp53_run/.

See the Workflows overview for the full output tree, common flags (GC bounds, top-N, verbosity), and Nextflow/off-target hand-offs.

3. Customize Inputs & References

Pass alternative sequences or transcriptomes to match your experimental data:

uv run sirnaforge workflow TP53 \
  --input-fasta examples/sample_transcripts.fasta \
  --transcriptome-fasta ensembl_mouse_cdna \
  --output-dir tp53_custom

  • --input-fasta accepts local paths or URLs, bypassing the transcript lookup while retaining familiar file naming via the positional gene argument.

  • --transcriptome-fasta selects the reference used for transcriptome off-target analysis (local/remote/preset). Provide it whenever you need species other than the bundled Ensembl set.

  • --offtarget-indices and --species feed the BWA-MEM2/Nextflow pipeline; details live in Workflows → Nextflow Pipeline.

4. Inspect Results

Key files in every workflow run:

File

Why it matters

sirnaforge/*_pass.csv

High-quality candidates ready for experiments

sirnaforge/*_all.csv

Full candidate list with all metrics for custom filtering

logs/workflow_summary.json

Summary of search/design stats, reference decisions, and QC flags

off_target/input_candidates.fasta

FASTA handed to the Nextflow off-target pipeline

Use standard CLI tools to browse:

head -6 tp53_run/sirnaforge/TP53_pass.csv
jq '.' tp53_run/logs/workflow_summary.json | less

For a complete walkthrough of directory contents, refer to Workflows → Output Structure.

5. Understand the Scores

siRNAforge surfaces the same thermodynamic metrics described in the code under sirnaforge/core/thermodynamics.py. Refer to:

Typical “green zone” checkpoints:

  • composite_score ≥ 8 for top-tier candidates

  • asymmetry_score ≥ 0.65 to ensure the guide strand loads into RISC

  • gc_content between 35–60% (optimal 40–55%)

6. Next Steps