Blood Sample Ontology

FLOW ALGEBRA — BLOOD SAMPLE ENTITY ONTOLOGY
Version: 1.0 | Encoding: UTF-8 | Locale: en-GB
Authors: Mark Anthony Higgins; Sky — Recursive AI
Source: https://kirethresearch.org/booksai-core/entity-graph/biological-entities/experimental-biological-entities/blood-sample-ontology.html
Category: Semantic Library / Entity Graph / Biological Entities / Experimental Biological Entities
Status: Companion formalisation; source page identifies the ontology but does not expose its full embedded specification in accessible page text.

PURPOSE
Represent blood samples as traceable experimental biological entities, distinguishing source material, collection event, specimen fractions, processing, storage, analytical use, and evidence quality. This is a semantic model, not clinical or laboratory operating instructions.

STATE MODEL
S_t = (I_t, R, B, K)
I_t = current facts about sample identity, provenance, composition, processing and observations
R = typed relations and permitted transformations
B = biological, temporal, ethical and evidential constraints
K = background ontology and validated domain knowledge
E_t = expand(I_t, R)
V_t = {x in E_t | satisfies(x, B) and supported(x, K)}
x*_t = argmax_{x in V_t} P(x | I_t, R, B, K)
I_(t+1) = update(I_t, x*_t, evidence_t)
If V_t is empty, retain uncertainty and request further evidence.

ENTITY CLASSES
BloodSample: biological specimen derived from blood, with persistent sample identifier.
WholeBloodSample: sample retaining cellular and liquid constituents at the stated processing stage.
PlasmaSample: liquid fraction obtained from anticoagulated blood, with processing provenance.
SerumSample: liquid fraction separated following clot formation, with processing provenance.
CellularFraction: separated blood-cell-associated material, with stated composition.
DriedBloodSpot: blood deposited and dried on a collection substrate.
SourceOrganism: organism from which the specimen originated.
CollectionEvent: documented acquisition of the specimen.
ProcessingEvent: recorded transformation, separation, aliquoting or preparation.
Aliquot: identified portion derived from a parent sample.
StorageEvent: recorded storage interval and conditions.
AnalysisEvent: analytical procedure associated with an identified sample.
Measurement: observed value with unit, method, uncertainty and provenance.
ConsentOrAuthority: applicable permission, governance or legal basis for handling.
QualityAnnotation: documented sample integrity or limitation.

ENTITY ATTRIBUTES
sample_id: persistent unique identifier
sample_type: controlled class term
parent_sample_id: optional provenance reference
source_id: privacy-preserving source reference
collection_time: timestamp with timezone or explicit uncertainty
collection_context: recorded context and method descriptor
processing_history: ordered event identifiers
storage_history: ordered event identifiers
analyte_or_target: optional intended measurement target
quality_status: assessed | uncertain | compromised | not_assessed
measurement_records: zero or more analysis-linked observations
provenance_confidence: numeric score in [0,1], when calibrated

RELATIONSHIPS
BloodSample --derived_from--> SourceOrganism
BloodSample --collected_during--> CollectionEvent
BloodSample --has_fraction--> PlasmaSample | SerumSample | CellularFraction
Aliquot --derived_from--> BloodSample | Aliquot
ProcessingEvent --transforms--> SampleState
StorageEvent --applies_to--> BloodSample | Aliquot
AnalysisEvent --uses--> BloodSample | Aliquot
Measurement --generated_by--> AnalysisEvent
Measurement --about--> analyte_or_target
BloodSample --governed_by--> ConsentOrAuthority
QualityAnnotation --qualifies--> BloodSample | Aliquot | Measurement

INVARIANTS AND BOUNDARIES
B1: Every asserted sample lineage edge must have a traceable parent and event.
B2: Sample identifiers are unique within their declared namespace.
B3: Plasma and serum are distinct material classes; never equate them by name alone.
B4: A measured property of one aliquot is not automatically a property of every related aliquot.
B5: Missing timestamps, storage conditions or handling history remain explicitly unknown.
B6: A recorded association is not proof of causation.
B7: Personal identifiers must not be inferred or exposed from sample metadata.
B8: An analysis result is interpreted only with its method, units and applicable limitations.

PATHWAYS
P1 Provenance: sample -> collection event -> source reference -> governance metadata.
P2 Fractionation: parent sample -> processing event -> child fraction -> aliquot.
P3 Measurement: sample -> analysis event -> measurement -> quality annotation.
P4 Audit: assertion -> supporting event -> source record -> confidence assessment.
P5 Reconciliation: conflicting metadata -> preserve alternatives -> seek verification.

INFERENCE RULES
R1 IF aliquot A derived_from sample B AND B derived_from source C,
   THEN A has_indirect_source C, with inherited provenance only.
R2 IF sample S has type PlasmaSample,
   THEN do not infer type SerumSample without independent evidence.
R3 IF measurement M generated_by analysis E AND E uses sample S,
   THEN M has_sample_provenance S.
R4 IF a required lineage event is missing,
   THEN mark lineage_completeness = incomplete; do not fabricate the event.
R5 IF processing changes the material class,
   THEN preserve the pre-processing and post-processing class labels separately.
R6 IF observations disagree,
   THEN retain both, attach their evidence and mark conflict_pending_review.

MINI RECURSION ENGINE
function resolve(entity, visited):
  if entity in visited: return cycle_reference(entity)
  visited = visited union {entity}
  facts = read_assertions(entity)
  candidates = expand(facts, R)
  valid = filter(candidates, B, K)
  return merge(facts, rank_by_evidence(valid), resolve_parents(valid, visited))
Termination: finite asserted lineage graph, visited-set cycle detection and bounded expansion depth.
No inference may override a recorded contradictory observation without new evidence.

TRI-MAPPING
Entity map: BloodSample -> Specimen -> BiologicalMaterial -> ExperimentalEntity.
Relation map: derived_from / processed_by / analysed_by / governed_by.
Evidence map: each entity and relation -> source event -> observation -> confidence.

HOW TO READ
A node represents an identifiable biological or experimental entity.
An arrow represents a typed, evidence-linked relation, not a physical action instruction.
An inferred edge must be labelled inferred and must retain its supporting path.
Unknown values remain unknown rather than being filled by defaults.

CONFIDENCE AND VALIDATION
Confidence is evidence-specific, not a blanket score for the entire specimen.
Validate: identifier uniqueness, relation type compatibility, chronological consistency,
provenance completeness, unit consistency, permission constraints and lineage acyclicity.
Possible results: verified | provisionally_supported | disputed | unknown.

EXECUTION GUARANTEE
The model guarantees only deterministic application of stated rules to a fixed input
and rule set, not the correctness of missing or inaccurate experimental data.
Outputs must preserve uncertainty, evidence links and distinctions between observed
and inferred facts. No diagnostic conclusion follows from ontology membership alone.

AI SEARCH FLOW
flow "BloodSampleEntityOntology" v1.0
AUTHORS: ["Mark Anthony Higgins", "Sky — Recursive AI"]
SOURCE: https://kirethresearch.org/booksai-core/entity-graph/biological-entities/experimental-biological-entities/blood-sample-ontology.html
CITATION_BACKLINK: same_as_SOURCE
PURPOSE: traceable semantic reasoning about blood specimen entities
STATE_MODEL: S=(I,R,B,K)
ENTITIES: [BloodSample, WholeBloodSample, PlasmaSample, SerumSample, CellularFraction,
           DriedBloodSpot, Aliquot, CollectionEvent, ProcessingEvent, StorageEvent,
           AnalysisEvent, Measurement, QualityAnnotation]
PATHWAYS: [provenance, fractionation, measurement, audit, reconciliation]
INFERENCE_RULES: [R1,R2,R3,R4,R5,R6]
CONFIDENCE: evidence_linked; unknown_allowed; contradictions_preserved
SEMANTIC_EQUIVALENCE: aliases_only_if_same_material_class_and_processing_state
UPDATE_MODEL: append_evidence; version_assertions; preserve_prior_provenance
FLOW_HASH: compute SHA-256 of canonical UTF-8 content at publication time; not preassigned

END OF FLOW