Ontogenesis — MIT-Licensed Ontology Genesis, Linking, and Alignment Framework

Base IRI: https://socioprophet.dev/ont/ontogenesis#
Current Version IRI: https://socioprophet.dev/ont/ontogenesis#v0.3.0

Ontogenesis is a reusable framework for constructing, linking, validating, and evolving bespoke ontological graphs under uncertainty.

This repository currently includes Human Digital Twin-oriented core terms and examples, but its purpose is broader than any single application. Ontogenesis provides the scaffolding for ontology creation, ontology-to-language and language-to-ontology interfaces, graph binding, provenance capture, policy-aware validation, and iterative refinement through annealing loops.

What Ontogenesis is for

Ontogenesis supports:

• seed ontology creation from observation, claims, measurements, and partial structure
• ontology linking across local, domain, and world models
• mapping and binding layers between heterogeneous schemas and vocabularies
• provenance, policy, and validation over evolving graph artifacts
• iterative ontology refinement under uncertainty
• ontology alignment in the wild, including noisy or partially compatible external vocabularies

Layered Mapping and Binding Use Cases

Lower ontology bindings

Lower-layer bindings connect close-to-source semantics into a usable graph. These may include:

• sensor and wearable observations
• human-facing measurements
• device telemetry
• local embodiment and state representations
• consent, provenance, and evidence attached to raw or derived observations

This layer is where local data first becomes semantically structured.

Middle ontology bindings

Middle-layer bindings connect operational and relational structure between local observations and broader world models. These may include:

• claims, annotations, and knowledge artifacts
• cases, tasks, workflows, and interventions
• policy decisions and evaluation events
• interaction models between agents, systems, and environments

This layer is where local structure becomes actionable and relational.

Upper ontology bindings

Upper-layer bindings connect local and middle-layer structures to broader environmental, conceptual, or world-model semantics. These may include:

• world and environment models
• domain reference ontologies
• temporal, spatial, organizational, or systemic frames
• higher-order constraints and shared semantic anchors

This layer is where bespoke graphs become interoperable with broader semantic systems.

Ontology alignment in the wild

Ontogenesis is also intended for messy real-world alignment work, including:

• reconciling competing vocabularies
• attaching SKOS concepts to formal classes
• mapping external schemas into local graph structures
• using JSON-LD contexts as boundary surfaces
• preserving provenance while progressively normalizing and linking terms
• supporting partial, provisional, or evolving alignments instead of pretending the world is tidy

Annealing and Lifecycle

Ontogenesis assumes that ontology work begins with incomplete knowledge.

A graph may begin as a seed formed from weak observations, provisional labels, noisy measurements, or incomplete claims. Through repeated refinement, that graph can be:

• seeded
• normalized
• linked
• trusted
• made actionable
• delivered

This repository uses validation and lifecycle gates so graph artifacts can become more grounded over time rather than being treated as perfectly known from the start.

Current Framework Surfaces

The repository includes:

• ontogenesis.ttl — core ontology (Turtle)
• skos/*.ttl — SKOS vocabularies and concept schemes
• shapes/*.ttl — SHACL constraints
• mappings/*.ttl — interoperability mappings such as PROV, FHIR, and IEML
• context.jsonld — JSON-LD boundary/context surface
• examples/*.ttl — example datasets
• tests/*.rq — SPARQL invariant tests
• capd/*.json — capability packaging metadata
• tools/validate.py — local validation entrypoint

Relationship to Applications

Ontogenesis is a framework, not a single domain application.

Applications such as Human Digital Twin may use Ontogenesis as an ontology and interface substrate, but the framework is intentionally broader: it exists to support the creation and evolution of bespoke ontological graphs for humans, digital agents, hybrid systems, and their environments.

Diagrams

Ontogenesis

An auditable, policy-gated, supply-chain–traceable Git repository for RDF/OWL/JSON-LD ontologies and semantic web assets used across the SocioProphet stack.

This repo is designed to support:

Install validation dependencies:

python3 -m venv .venv && . .venv/bin/activate && pip install -r requirements.txt

Validate ontology + examples:

python -m venv .venv && source .venv/bin/activate
pip install -r requirements-dev.txt

python scripts/validate_rdf.py
python scripts/shacl_gate.py
python scripts/jsonld_roundtrip.py

python scripts/build_dist.py
python scripts/ledger_build.py
python scripts/ledger_verify.py

License

This repository is licensed under the MIT License. See LICENSE.

Release discipline (CI)

On a tag v*, CI will:

1. Build dist/ deterministically (SOURCE_DATE_EPOCH fixed).
2. Validate RDF + SHACL gates.
3. Build/verify ledger/ledger.csv.
4. COSE-sign each dist/* and record signature URIs in the ledger.
5. Emit a minimal SPDX SBOM (sbom/spdx.json) with checksums.

Policy

• dist/ and audit/ are generated only.
• Changes under Upper/, Middle/, Lower/, Domains/, Platform/, prophet/, epi/ must be accompanied by scripts/build_dist.py + scripts/ledger_build.py output changes.
• Promotion gates are enforced via SHACL bundles in shapes/.

See docs/ for the design plan and module map.