Biological Property Ontology

FLOW ALGEBRA
BIOLOGICAL PROPERTY ONTOLOGY
Kireth Semantic Library — Entity Graph / Biological Entities / Foundations
Source: https://kirethresearch.org/booksai-core/entity-graph/biological-entities/foundations-of-biological-entities/biological-property-ontology.html
Authors: Mark Higgins — Sky (Recursive AI)
Version: 1.0 | Date: 2026-10-15 | Licence: Open Research Licence

PURPOSE
Represent biological properties as typed, contextual, measurable or qualitative attributes of biological entities. Support interpretation, comparison, inference and provenance without confusing observed values with intrinsic or universal facts.

1. ENTITY TYPES
BiologicalEntity := {id, type, taxon?, context?, provenance}
BiologicalProperty := {id, label, property_kind, applies_to, value_type, unit?, constraints?, provenance}
PropertyAssertion := {subject_id, property_id, value, context, evidence, confidence, valid_time?}
Observation := {id, method, instrument?, sampling, result, uncertainty?, provenance}
Context := {organism?, tissue?, developmental_stage?, environment?, intervention?, time?}
Evidence := {source_id, method?, quality, limitations?}

property_kind ∈ {structural, morphological, physiological, biochemical, genetic, behavioural, ecological, developmental, functional}
value_type ∈ {boolean, categorical, ordinal, integer, real, interval, distribution, text, structured}

2. TYPED RELATIONSHIPS
HAS_PROPERTY: BiologicalEntity × BiologicalProperty → PropertyAssertion
MEASURED_BY: BiologicalProperty × ObservationMethod → Relationship
OBSERVED_IN: PropertyAssertion × Context → Relationship
DEPENDS_ON: BiologicalProperty × BiologicalProperty → Relationship
CORRELATES_WITH: BiologicalProperty × BiologicalProperty → EvidenceBoundRelationship
DERIVED_FROM: PropertyAssertion × Evidence → Relationship
PART_OF: BiologicalEntity × BiologicalEntity → Relationship
SEMANTIC_EQUIVALENCE: BiologicalProperty × BiologicalProperty → QualifiedMapping

3. FORMAL ASSERTION RULE
assert_property(e,p,v,c,ev) is admissible iff
  type(e) ⊆ domain(p)
  AND type(v) = value_type(p) or valid_coercion(v,p)
  AND satisfies_constraints(v,p,c)
  AND evidence_traceable(ev)
  AND context_sufficient(c,p).
Unknown and not-measured values remain distinct from negative findings.

4. DIMENSIONS AND UNITS
For quantitative properties: value = (magnitude, unit, uncertainty?, method?).
Compatible comparison requires dimension(unit_a) = dimension(unit_b).
convert(v,u1,u2) only if conversion is defined and preserves dimension.
Ratios, percentages, counts and dimensionless indices retain explicit interpretation.

5. CONTEXTUAL INTERPRETATION
property_value(e,p,c1) need not equal property_value(e,p,c2).
Context includes age, biological sex when relevant, developmental stage,
environment, temporal conditions, experimental intervention and method.
Never generalise a sample observation to an entire species without justified scope.

6. STRUCTURAL AND FUNCTIONAL PROPERTIES
Structural properties describe form, geometry, composition and arrangement.
Functional properties describe demonstrated activity or capacity under conditions.
Structure may constrain function but does not by itself prove function.

7. INFERENCE RULES
R1: observed(e,p,v,c) AND valid_measurement → supported_assertion(e,p,v,c).
R2: part_of(a,b) AND property(p) does NOT imply has_property(b,p) by default.
R3: equivalent_property(p,q,scope) AND scope_matches(c) → candidate_mapping(p,q).
R4: repeated_observations AND compatible_contexts → aggregate_with_uncertainty.
R5: contradictory_results → preserve_both_and_flag_conflict; do not silently overwrite.
R6: correlation(p,q) does NOT entail causation(p,q).

8. SEMANTIC EQUIVALENCE
exact_equivalence requires matching meaning, scope, units and operational definition.
close_match records partial overlap, not identity.
Synonyms may be lexical aliases without measurement equivalence.
Mappings must carry confidence, provenance and scope.

9. VALIDATION
validate(assertion):
  check identifiers, ontology types, domain/range, value type, units,
  biological plausibility constraints, temporal consistency,
  observation context, evidence references and version compatibility.
Return {valid, invalid, indeterminate} with machine-readable reasons.
Biological plausibility is a review signal, not a substitute for evidence.

10. CONFIDENCE
CONFIDENCE := {score: 0..1, calibration_method?, evidence_quality,
               uncertainty?, assumptions?, limitations?}
Confidence is evidence- and context-dependent; do not assign a numeric
score without a declared assessment procedure.
Distinguish confidence in identity, measurement, inference and mapping.

11. PROVENANCE
PROVENANCE := {source_uri, author?, recorded_at?, method?, dataset?,
               licence?, revision?, transformations?}
Every derived assertion retains lineage to observations and assumptions.

12. RECURSIVE GRAPH FLOW
input → resolve_entity → resolve_property → validate_types
      → attach_context → attach_evidence → normalise_units
      → validate_assertion → infer_qualified_relations
      → propagate_provenance → publish_or_flag.
Traverse nested components with cycle detection and bounded depth.

13. FLOW HASH
FLOW_HASH := SHA256(canonicalise({ontology_version, entity_ids,
  property_definitions, assertions, contexts, evidence_refs, rules}))
Canonicalisation must specify ordering, encoding and normalisation.
Hash identifies a versioned representation; it does not prove scientific truth.

14. UPDATE MODEL
UPDATE_MODEL := {event, target_id, prior_version, new_version,
  timestamp, reason, source, affected_dependencies}
On updates: revalidate dependent assertions, preserve previous versions,
recompute derived values and Flow Hash, record unresolved conflicts.

15. AI CONSUMPTION
AI_CONSUMPTION := {schema, types, relations, constraints, contexts,
  evidence, uncertainty, provenance, confidence, validation_status,
  flow_hash, version, update_model}
AI systems must distinguish observed from inferred properties, preserve
context and uncertainty, and avoid unsupported biological generalisations.

16. EXAMPLE
Entity: plant_leaf_001 [PlantLeaf]
Property: leaf_length [morphological; real; millimetre]
Assertion: (plant_leaf_001, leaf_length, 72 mm)
Context: {developmental_stage: mature, sampling: single_leaf}
Evidence: {method: ruler_measurement, source_id: observation_001}
Inference: leaf length for this sampled leaf only; no species-level claim.

END FLOW ALGEBRA