Skip to content

Roadmap

Big Picture

Visualization Studio

Phase 1: Foundation ✅

Interactive dashboards with modern web components and real-time data binding

Data Ingestion

Phase 1: Foundation ✅

Multiple tabular formats with Delta Lake storage and data provenance

MultiQC Integration

Phase 2: Specialized Components ✅

Seamless integration with bioinformatics quality control reports

step icon

Image Grid Component

Phase 2: Specialized Components ✅

S3-backed image galleries with configurable thumbnails and grid layout

Geospatial Map Component

Phase 2: Specialized Components ✅

Scatter, density, and choropleth maps with cross-filtering and GeoJSON / Map-capable Table DCs (v0.12.0)

Advanced Biology Visualizations

Phase 2: Specialized Components ✅

Omics plots extending the v0.12 DC-level type-config pattern — Volcano, Manhattan, Sunburst, and more

Templates & Community

Phase 3: Templates ✅

Reusable dashboard templates for standard bioinformatics workflows, nf-core integration, and nf-core plugin for automatic data ingestion

step icon

React Viewer

Phase 3.5: UI Modernization ✅

Vite + Mantine SPA — Dash removed in v0.13.12, canonical URLs live in v1.0.0

Citable Science

Phase 4: Reproducibility 📋

DOI-backed snapshots and sample-to-viz provenance on top of the existing Serve hosting

AI & Intelligence

Phase 5: Intelligence 📋

Smart dashboard creation & AI-assisted analysis

Reproducibility & FAIR

Built with FAIR principles in mind. v0.12 ships YAML-defined dashboards, Cross-DC linking, nf-core/ampliseq templates, and hosting via SciLifeLab Serve ✅. Citable DOI snapshots, persistent IDs, and data lineage land in Phase 4.

Challenge Status
Reproducibility requires setup nf-core/ampliseq templates shipped (docs); nf-core plugin planned
Can't reproduce visualizations YAML-defined dashboards + traceable data (docs)
Siloed experiment data Cross-DC linking (docs)
Dashboards disappear Hosted on SciLifeLab Serve (changelog)
Not citable / no sample-to-viz traceability Phase 4 — DOI snapshots via SciLifeLab Serve + LabID provenance
No data lineage Phase 4 — Delta Lake time travel, auditing

Current Features

Data Ingestion

  • YAML-based data ingestion via CLI (docs)
  • Polars-compatible formats (Parquet, CSV, JSON, TSV) → Delta Lake
  • S3/MinIO storage with backup/restore commands (docs)
  • MultiQC report integration (docs | #626)
  • MultiQC data lifecycle — append / replace / clear runs from the viewer with uniformity validation (v0.12.0, docs)
  • DC-level type configuration — Map-capable Table DCs (lat/lon column detection on upload, DCTableCoordinatesConfig); extensible to advanced-viz types (v0.12.0, docs)
  • Client-side table joining in CLI ( #634)
  • Recipe-based data transformation — Python recipes with 4-checkpoint validation (docs)
  • Template-based project setup — one-command project creation with {DATA_ROOT} substitution (docs)

Dashboard Components

Specialized components share a common pattern: a DC-level type-config (introduced for Map in v0.12.0) declares the columns that drive the visualization, so every figure built on top inherits the config — no per-figure boilerplate. Advanced biology visualizations extend the same pattern.

  • Generic components: Figure, Table, Card, Interactive (docs)
  • MultiQC components for QC reports (docs)
  • Image grid with S3/MinIO integration and configurable thumbnails ( #664)
  • Geospatial map component: scatter, density, choropleth with GeoJSON DC support + Map-capable Table DCs (v0.12.0, docs)
  • Figure code mode with live preview ( #639)

Dashboard Interactivity

  • Two-panel layout with tabs ( #616)
  • Cross-DC filtering via universal linking (docs)
  • Cross-DC links UI — Create / Edit / Delete links from the React Beta viewer with resolver picker and sample-mapping preview (v0.12.0, docs)
  • Interactive selection filtering: scatter/table/map selections
  • YAML dashboard import/export (docs)

Infrastructure

  • Docker + Kubernetes with Helm charts (docs)
  • SciLifeLab Serve hosting via values-serve.yaml overlay (changelog)
  • Celery/Redis background processing
  • Authentication: local, Google OAuth, unauthenticated mode

Frontend

  • React viewer — Vite + Mantine SPA on canonical URLs (v1.0.0); Dash removed in v0.13.12

Planned Features

Phase 3: Templates & Community

Reusable dashboards for standard bioinformatics workflows, with a focus on nf-core community adoption.

Infrastructure (shipped)

  • Depictio templates system — one-command project setup via depictio run --template, with {DATA_ROOT} substitution, template provenance tracking, and automatic dashboard import (docs)
  • Recipe-based data transformation — versioned Python recipes with 4-checkpoint validation, co-located with templates (docs)

Planned

  • nf-core plugin — automatically registers nf-core pipeline outputs in Depictio at run time, no manual CLI step required
  • nextflow.config template — embed Depictio ingestion directly in your Nextflow config so data collection happens at pipeline runtime
  • Snakemake report plugin — Depictio as a drop-in replacement for Snakemake's built-in HTML report, with interactive dashboards instead of static outputs

Phase 3.5: React viewer — v1.0.0 ✅

The React viewer (Vite + Mantine SPA) is the sole frontend as of v1.0.0. Dash was removed in v0.13.12; canonical URLs went live in v1.0.0.

  • React viewer shipped (v0.12.0)
  • Dash frontend removed (v0.13.12)
  • URL graduation — canonical /dashboards, /dashboard/{id}, /dashboard-edit/{id} (v1.0.0)
  • Component parity audit complete

Phase 4: Citable Science

DOI-backed citability and sample-to-viz provenance on top of the existing SciLifeLab Serve hosting.

  • DOI snapshots — persistent stable URLs per dashboard version, citable in publications
  • Sample-to-viz provenance — LabID linking sample → pipeline run → Delta table → visualization
  • Static export — Quarto for publication supplements

Phase 5: AI & Intelligence (12+ months)

  • Smart dashboard creation — describe the analysis you need; AI proposes a layout and component configuration
  • AI-assisted analysis — automated anomaly detection, visualization recommendations, and plain-language report narration
  • MCP server — expose Depictio as a tool for AI agents to create dashboards, query data, and manage projects programmatically