Helix StudioGenome IDE
← Docs
Helix CLI docs
Browse Helix CLI docs

Helix

Helix is a research‑grade bioinformatics toolbox that “proves what it plots.” Every figure and workflow carries schema validation, versioned specs, and cryptographic provenance so you can trust what you share and reproduce what you see.

  • Install: pip install "helix-governance[viz,schema,protein]" (3.12 tested)
  • First look: helix demo viz --all
  • Why it matters → Artifacts & Provenance

Governance licensing

Helix is free to run (simulate + explore). Governance/decision-grade assertions and signed bundles require an offline license file. Verification is always free + offline: helix verify <bundle_dir_or_zip>.

Quickstart

Run a triage snapshot (GC skew + k‑mer hotspots + ORFs) and save a ready‑to‑share plot:

helix triage --input path/to/sequence.fna --k 5 --min-orf-length 90 --json triage.json
helix viz triage --json triage.json --output triage.png

Fold RNA and export dot‑bracket and ensemble plots:

helix rna mfe --fasta path/to/seq.fna --dotbracket mfe.dbn
helix rna ensemble --fasta path/to/seq.fna --gamma 1.0 \
  --dotplot dotplot.png --arc arc.png --entropy-plot entropy.png

Notebook‑friendly API (plain dict/list payloads):

from helix import api as hx
report = hx.triage_report(sequence="AUGGCCUUUUAA", k=3)
fold   = hx.fold_rna("GGGAAACCC", min_loop_length=0)

Interactive edit-DAG demos

Two end-to-end recipes ship in examples/ so you can go from FASTA/configs → DAG → Playground in seconds:

  • CRISPR demo (guide library)

    helix crispr dag \
  --genome examples/hg19_chr_demo.fa \
  --cas-config examples/cas9.json \
  --guides-file examples/guides.tsv \
  --region chrDemo:1-500 \
  --out-dir out/crispr_dags/

    Artifacts appear in out/crispr_dags/ and a prebuilt copy lives at data/crispr_demo.edit_dag.json. Open in Playground

  • Prime editing demo (peg sweep)

    helix prime dag \
  --genome examples/hg19_chr_demo.fa \
  --editor-config examples/prime_editor.json \
  --pegs-file examples/pegs.tsv \
  --region chrDemo:1-500 \
  --out-dir out/prime_dags/

    Produces helix.prime.edit_dag.v1.1 artifacts per peg (see data/prime_demo.edit_dag.json for a hosted example). Open in Playground

  • Experiment configs

    helix experiment new --type crispr --out experiments/demo_crispr.helix.yml
helix experiment run --config experiments/demo_crispr.helix.yml --out out/demo_crispr.edit_dag.json
helix experiment report --config experiments/demo_crispr.helix.yml --out out/demo_crispr.html

    A single YAML file describes the entire “edit experiment” (genome, Cas/Prime configs, guides/pegs, simulation knobs). Helix can regenerate DAGs, PNGs, and HTML reports from that spec at any time; see templates/*.helix.yml for starter files.

  • HelixSpec run pipeline (compile → lower → execute → verify): see HelixSpec quick start.

Live realtime Playground

  • Realtime CRISPR Simulator — explore guides, stream DAG frames, and compare two designs interactively (browser-only, no backend required).
    • Suggested demo: Guide A ACCCAGGAAACCCGGGTTTT vs Guide B TTTACCCAGGAAACCCGGGT (chrDemo) to see how probability mass shifts live.
    • Branch chart + entropy panel update every frame (intended, indel, no-edit) so you can watch probability flow in real time.
    • Export frames JSONL and hand them to helix edit-dag generate-dataset --frames-input … to append real experiments to your ML corpora (see Edit DAG Frames).
    • Prefer a desktop shell? pip install helix-governance[gui] && helix gui launches the PySide6 version with the same streaming frames and Cytoscape canvas.

Deep Wiki (Project Map)

Use this “deep wiki” map to jump directly to the concept or tool you need:

Benchmarks & Reproducibility

Helix includes a research‑grade benchmark harness. Every run emits a schema‑tagged JSON that captures git SHA, BLAS vendor, CPU/threads, RNG seed, per‑case timing, and RSS peaks. CI appends results to data/bench/history.csv; the live dashboard renders trends without servers or external deps.

  • See live trends → Benchmarks
  • Harness CLI → python -m benchmarks.api_benchmarks --help

Design Principles

  • Evidence‑first: every artifact ships with a contract (schema) and provenance (hashes, spec version, parameters).
  • Friendly internals: approachable implementations of classic algorithms, with JSON payloads you can inspect and remix.
  • CLI ⇄ API symmetry: what you can do from the shell, you can reproduce in a notebook—and vice versa.
  • Research‑grade: optional heavy datasets, performance dashboards, and drift gates in CI for trustworthy iterations.