Everything your lab needs, in one place

From raw sequencing data to publication-ready results, bioAF brings pipelines, interactive analysis, experiment tracking, and cost control together in one platform, on your own cloud. Here is what is inside.

Get Started

Pipeline Engine

Run bioinformatics workflows without managing compute infrastructure, and author your own when the catalog isn't enough.

  • A live, searchable library of every nf-core pipeline: browse the full registry, pick a version, and install in one click
  • One-click launch with parameter overrides
  • Real-time DAG visualization and log streaming per stage
  • Custom pipelines run any program that runs on Linux: bash, Python, R, or arbitrary tools
  • Develop in Work Nodes, then promote to a versioned pipeline backed by a GitHub repo or pasted script
  • Custom QC dashboards driven by your own metrics, with full provenance from inputs to outputs
Deep dive →

Notebook Sessions & Work Nodes

Interactive analysis in your browser, or a full Linux VM when you need to bring your own tools.

  • JupyterHub for Python, RStudio Server for R
  • Work Nodes: full Linux VMs with SSH, conda environments, and your GitHub repos pre-cloned
  • Versioned environments built from a Dockerfile or conda spec
  • Auto-stop on idle to eliminate unnecessary cloud spend
Deep dive →

Experiment Management

Track experiments from bench to publication with structured, standards-compliant metadata.

  • MINSEQE-compliant metadata for every experiment
  • Sample batching with a defined status lifecycle
  • Files, pipeline runs, and results all linked back to their source experiment
  • GEO submission export with required fields pre-populated
Deep dive →

Results & Visualization

Explore and share analysis results, from single-cell embeddings to QC reports.

  • Integrated CellxGene Discover for single-cell data assembly and browsing
  • Auto-generated QC dashboards after every pipeline run
  • Searchable plot archive with full-resolution downloads
  • Role-based sharing so PIs can review results without touching the pipeline
Deep dive →

Literature Library

A shared paper library with multi-source search and in-app reading, plus AI Literature Review that finds relevant new papers for each experiment.

  • AI Literature Review: the LLM expands your search, scores candidates for relevance, and adds top papers to the library, associated with the experiment
  • Run it on demand, or on an automated daily, weekly, or monthly cadence for experiments with new activity
  • Search PubMed, bioRxiv, Europe PMC, and Semantic Scholar; results join the library only when you choose to keep them
  • Upload PDFs with automatic title, author, DOI, and abstract extraction, and read them in a paginated in-app viewer
  • Threaded comments, reading-status tracking, and experiment or project associations
  • Curated papers feed back into AI Review, which flags results that contradict prior work
  • Export any paper, or a whole filter, as BibTeX or RIS
Deep dive →

AI Review

Optional, opt-in LLM advisories on completed pipeline runs and full experiments. Bring your own API key.

  • Three providers supported: Anthropic Claude, OpenAI ChatGPT, and Google Gemini
  • Per-org admin chooses one active provider and model; switching is one click
  • Severity-coded cards (red, orange, green) on a new AI Review tab
  • Section-based prompt builder, with org-wide saved prompts for repeated questions
  • Experiment-wide reviews stale-mark themselves when a new run is added
  • Never sends raw reads, file bytes, or pipeline logs. Every invocation is audit-logged
Deep dive →

Data Management

Centralized file management with automatic organization across the research lifecycle.

  • Upload, browse, and download experiment data in one place
  • Four-tier storage lifecycle: Ingest, Raw, Working, Results
  • Auto-ingest for sequencer or CRO output files via Cloud Pub/Sub
  • All data stays in your own GCP project: no hosted dependencies
Deep dive →

Cost Center

Real-time cloud spending visibility without needing to navigate the GCP console.

  • Components auto-scale to control cost when not in use
  • Per-component and per-project spend breakdowns
  • Trend charts with month-to-date projections
  • Budget alerts at 50%, 80%, and 100% of your threshold
  • Idle resource detection and cost optimization recommendations
Deep dive →

Infrastructure Management

Enable and disable platform components from the UI, no Terraform or Kubernetes experience needed.

  • Component catalog with plain-language descriptions and cost estimates
  • Preview the infrastructure plan before anything is provisioned
  • Version-controlled infrastructure history with one-click rollback
  • In-app upgrade notifications: review the changelog before applying
Deep dive →

LIMS Integration API

A public, key-authenticated REST surface so your LIMS and other lab systems read and write bioAF records without scraping the UI.

  • Projects, experiments, samples, and file metadata at /api/v1/integrations/*
  • Service accounts with role-scoped API keys, minted from the admin UI
  • Look up records by your own LIMS-side IDs via external_id
  • Idempotency keys on writes so retries never duplicate records
  • Signed outbound webhooks for experiment, sample, and file events with retries and a dead-letter queue
  • Authoritative OpenAPI schema served by every instance at /api/v1/integrations/openapi.json
Deep dive →

Audit & Compliance

An immutable record of everything that happened, who did what, when, and to which resource.

  • Every experiment, pipeline run, session, and file event is logged
  • Human-readable descriptions, not raw event codes
  • Trace any result back to its exact input data and parameters
  • Exportable for regulatory reviews and compliance reporting
Deep dive →

Ready to get started?

bioAF is free and open source. Deploy it on your own cloud account in under an hour.

Read the Setup Guide