Skip to content

Platform Overview

1. Platform Overview & Vision

Document status: Draft v0.1 Last updated: 2026-03-31 Depends on: Component sec1 docs (Hippo, Cappella, Canon, Aperture, Bridge) Feeds into: sec2_components.md, sec3_integration.md, deployment docs, getting-started guide

1.1 What BASS Is

BASS (Bioinformatics Analysis Software System) is an open-source, modular platform for managing and analyzing biological research data. It provides a unified foundation for:

  • Metadata management — a single, queryable registry of entities (subjects, samples, datafiles, analysis outputs) with full provenance
  • Artifact resolution — deterministic, reproducible resolution of where canonical output files live for any set of input entities
  • Data harmonization — automated ingestion and consistency enforcement across external laboratory information systems (LIMS), clinical databases, and analysis pipelines
  • Access control — optional authentication and authorization for shared deployments

BASS is designed for research labs and institutions working with high-dimensional biological data: genomics, transcriptomics, spatial biology, clinical omics, and related fields. It targets the gap between raw data files and downstream analyses — the metadata and logistics layer that is often managed via spreadsheets, ad hoc scripts, and institutional inertia.

1.2 Design Principles

1.2.1 Modularity — Use What You Need

BASS components are independently installable and usable. A researcher who only needs metadata tracking can install Hippo alone; a team that additionally needs file artifact resolution adds Canon; a lab with external LIMS systems adds Cappella for integration; a multi-user deployment adds Bridge for auth.

No component requires another as a runtime dependency. Integration is opt-in.

1.2.2 SDK-First Architecture

All business logic lives in Python SDKs. REST APIs and GraphQL (future) are thin transport wrappers over the same SDK logic. This means:

  • Local use (laptop, notebook) is as capable as server-mode use
  • The REST API does not expose capabilities that the SDK does not
  • Components can be used programmatically without running a server

1.2.3 Progressive Deployment

The same codebase runs from a single researcher's laptop to an institutional cloud:

Scale Components Auth Storage
Individual researcher Hippo + Canon SDK only None SQLite
Small lab Hippo + Cappella + Aperture CLI None (or Bridge API keys) SQLite
Multi-user team All components + Bridge Bridge API keys SQLite or PostgreSQL
Institution All components + Bridge Bridge OIDC PostgreSQL

Scale is controlled entirely by configuration. No code changes required to move between tiers.

1.2.4 Provenance by Default

Every data mutation in Hippo records who changed what, when, and via which path. This is not an add-on — it is fundamental to the data model. Provenance enables:

  • Reproducibility: trace the origin of any entity back to its source
  • Debugging: find when and how an incorrect value was introduced
  • Auditing: demonstrate to collaborators or reviewers what changed in the dataset

1.2.5 Config-Driven Domain Adaptation

BASS is domain-agnostic at the platform level. Entity types, fields, relationships, and validation rules are defined by deployment-specific configuration files (schema.yaml, validators.yaml). The same BASS codebase runs for a genomics lab, a cell therapy program, a clinical biobank, or a plant biology consortium — with different configurations.

1.3 Platform Scope

In Scope

Capability Primary Component
Entity metadata storage and querying Hippo
Data provenance and audit trail Hippo
File artifact resolution Canon
External system ingestion (LIMS, REDCap, CSV) Cappella
Pipeline execution and output ingestion Cappella
Command-line interface for researchers Aperture
Multi-user authentication and access control Bridge
Cross-component data consistency Bridge

Out of Scope (v1.0)

Capability Status Notes
Bioinformatics analysis execution Out of scope BASS manages data about analyses, not analyses themselves (except file cache in Canon)
File storage and data transfer Out of scope BASS stores paths/URIs, not file contents
Web portal (beyond CLI) Deferred to v1.1+ Aperture web portal in design; CLI ships in v1.0
GraphQL API Deferred to v1.1 REST first; GraphQL gateway via Bridge in v1.1
Multi-institution federation Deferred to v2.0 Bridge as inter-BASS federation layer — needs design
Full RBAC and OAuth 2.0 Deferred to v1.1 v1.0 ships API key auth only

1.4 Relationship to Existing LIMS Systems

BASS is not a replacement for existing LIMS systems (STARLIMS, REDCap, LabKey, etc.). It is a metadata intelligence layer that aggregates and harmonizes data from those systems.

Existing LIMS systems continue to be the system of record for wet-lab operations (sample tracking, accessioning, clinical data entry). Cappella pulls from those systems and harmonizes the results into Hippo's canonical schema. Researchers query Hippo for analysis rather than querying multiple upstream systems directly.

STARLIMS   REDCap   Clinical DB   Sequencing Core
    │          │          │              │
    └──────────┴──────────┴──────────────┘
                          │
                     Cappella
                    (harmonize)
                          │
                        Hippo
                    (canonical store)
                          │
                Canon + Aperture
                  (query + resolve)

1.5 v1.0 Scope and Release Criteria

Platform v1.0 is the first production-ready release of the full component stack.

v1.0 components: - Hippo v0.4+ (stable metadata store, REST API, provenance, schema migrations) - Canon v0.3+ (artifact resolution, CWL execution integration, storage adapters) - Cappella v0.2+ (trigger engine, external adapters, reconciliation) - Aperture v0.1 (CLI: entity CRUD, schema inspection, provenance, auth commands) - Bridge v0.1 (API key auth, request routing, cross-component sync)

v1.0 exit criteria: - A new user can install, configure, and complete the getting-started walkthrough using only the documentation - Full integration test suite (round-trip: external source → Cappella → Hippo → Canon) passes in CI - Docker Compose single-node deployment works from a clean machine - All components have tagged releases and documented changelogs

1.6 Open Questions

Question Priority Status
Platform name: "BASS" vs. "DryLIMS" vs. other Low Open — revisit before v1.0 public announcement
Multi-institution federation model Medium Candidate: Bridge as federation gateway; needs design for v2.0
GraphQL gateway scope for v1.1 Low Open