Lineage Graph

Lineage Graph is a feature of the Code Ocean computational science platform that provides an automated, immutable record of how result data is generated. It captures the full provenance of every result — from source data through processing steps in Compute Capsules and Pipelines, all the way to the final output — giving computational scientists, R&D leaders, and IT teams complete visibility into their research workflows.

Designed for life sciences and computational biology teams, Lineage Graph ensures that every piece of result data can be traced, inspected, and assessed for reproducibility. It is tightly integrated with the broader Code Ocean platform, working seamlessly alongside Compute Capsules, Pipelines, and Data management to deliver end-to-end traceability without requiring manual documentation.

Key Capabilities

• Immutable result tracking: All computational work is tracked in Git, creating a permanent, tamper-proof record of every analysis performed and the results it produced. Users can click into any result to access key details at a glance.
• Full result provenance exploration: The Lineage Graph exposes the entire history of how a result was produced. Users can enter any node in the graph to access the exact version and Git commit of the Capsule or Pipeline responsible for generating each result.
• Reproducibility assessment: The Lineage Graph evaluates the reproducibility of any given result. Reproducibility is guaranteed when Capsules and Pipelines are versioned and when data is stored internally on the platform. If external data sources subject to change are used, reproducibility cannot be guaranteed.

How the Lineage Graph Works with the Code Ocean Platform

• Compute Capsules: All computational work performed within Compute Capsules is automatically recorded. Results generated are captured along with the input data used to create them, ensuring a complete audit trail.
• Pipelines: Pipelines can incorporate output from multiple Compute Capsules, data sources, and other Pipelines. The Lineage Graph tracks all of these connections, providing a unified view of complex, multi-step workflows.
• Data tracking: Code Ocean tracks all data used in computational work. Data and results are accessible through the Data dashboard, enabling users to see exactly what data was used, where, and when.

Interactive Graph Features

• Every node on the Lineage Graph is interactive — users can click on any asset they have access to and explore full details including which version was run, when it was run, and by whom.
• Via the Code Ocean API, the Lineage Graph can be connected to external lineage tools, such as open-source solutions like OpenLineage, enabling integration with existing data governance infrastructure.

Supported Use Cases

• Data analysis: Supports reproducible data analysis workflows using ready-made Compute Capsule templates, any preferred language or IDE, open-source software, and built-in containerization.
• Data management: Manages organisational data in compliance with FAIR principles, using custom metadata and controlled vocabularies for consistency and searchability.
• Bioinformatics pipelines: Supports building, configuring, and monitoring bioinformatics pipelines with a visual builder or one-click nf-core import, running on AWS Batch with automatic scaling.
• ML model development: Provides GPU-ready environments and MLflow integration for developing, tracking, and managing ML, deep learning, and generative AI models, with built-in reproducibility and lineage.
• Multiomics: Handles large multimodal datasets with scalable compute and storage, cached R and Python packages, and preloaded multiomics analysis software, all with full lineage tracking.
• Imaging: Supports image processing workflows ranging from desktop applications to custom deep learning pipelines, across datasets from individual files to petabyte scale, always with lineage.
• Data and model provenance: Provides automated result provenance and lineage graph generation with a visual representation of every Capsule, Pipeline, and Data asset involved in a computation.

Lineage Graph is part of the Code Ocean cloud platform, which also includes Compute Capsules, Pipelines, Data management, ML Models, Collections, Apps, and an Admin Panel. The platform supports programmatic access via API and is built to meet the needs of computational scientists, bench scientists, IT and engineering teams, and R&D leadership in life sciences organisations.