Microbial Sample Ontology

FLOW ALGEBRA — MICROBIAL 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/microbial-sample-ontology.html
Category: Semantic Library > Entity Graph > Biological Entities > Experimental Biological Entities
Status: Companion formalisation; original embedded page code was not accessible in extracted page text.
Purpose: Machine-readable conceptual reasoning over microbial samples, their origins, compositions, handling histories, observations and evidence.

1. FLOW DECLARATION
flow "MicrobialSampleEntityOntology" v1.0
  implements [BiologicalSpecimen, Sample, ExperimentalEntity, Provenance, Evidence, QualityControl]
  input: query, sample_record, observations?, context?
  output: entity_graph, classifications, inferred_relations, confidence, provenance, uncertainty

2. STATE MODEL
S_t = (I_t, R, B, K)
I_t: currently known sample facts and observation records
R: typed relationships and inference rules
B: admissibility boundaries (identity, provenance, method, time, evidence)
K: domain knowledge, controlled vocabulary, reference metadata
E_t = expand(I_t, R)
V_t = {x in E_t | satisfies(x, B)}
x* = argmax_{x in V_t} P(x | I_t, R, B, K)
I_(t+1) = I_t union validated(x*)
Stop when no new justified facts can be added or an uncertainty boundary is reached.

3. ENTITY TYPES
MicrobialSample: a collected portion of material potentially containing microorganisms or microbial components.
SourceMaterial: environmental, clinical, food, water, soil, surface, host-associated or other source.
CollectionEvent: time, location category, method reference and collector provenance.
SampleContainer: identifier and containment metadata.
SampleAliquot: derived portion retaining a parent-child link.
MicrobialCommunity: potentially multiple taxa or populations in one sample.
MicrobialIsolate: culture-derived population represented as a distinct entity; not equivalent to the original mixed sample.
MicrobialTaxon: taxonomic classification with reference and version.
MolecularObservation: sequence, marker or other molecular result.
PhenotypicObservation: measured or described characteristic with method and conditions.
QualityRecord: integrity, contamination flags, chain of custody and controls.
ProcessingEvent: recorded transformation with input, output and timestamp.
AnalysisResult: interpreted result linked to evidence and analytical method.

4. CLASSIFICATIONS
sample_basis = {environmental, host_associated, food, water, soil, surface, laboratory_reference, other, unknown}
composition = {single_taxon_supported, mixed_community_supported, undetermined}
material_state = {raw, preserved, processed, aliquoted, archived, discarded, unknown}
identity_status = {verified, provisional, disputed, unknown}
quality_status = {acceptable, flagged, invalid, unassessed}
No category alone proves viability, purity, pathogenicity or taxonomic identity.

5. RELATIONSHIP GRAPH
MicrobialSample --derived_from--> SourceMaterial
MicrobialSample --collected_during--> CollectionEvent
MicrobialSample --held_in--> SampleContainer
SampleAliquot --derived_from--> MicrobialSample
MicrobialSample --may_contain--> MicrobialCommunity
MicrobialIsolate --derived_from--> MicrobialSample
Observation --observes--> MicrobialSample | MicrobialIsolate | SampleAliquot
Observation --uses_method--> MethodReference
AnalysisResult --supported_by--> Observation
AnalysisResult --assigns_taxon--> MicrobialTaxon
ProcessingEvent --consumes--> SampleEntity
ProcessingEvent --produces--> SampleEntity
QualityRecord --assesses--> SampleEntity
All edges retain evidence identifiers and timestamps where applicable.

6. INFERENCE RULES
R1: If an aliquot is derived from a parent sample, inherit source provenance, but do not automatically inherit measured composition or quality.
R2: If a taxonomic label is supported by a documented analysis, attach it as an evidence-qualified assignment, not an unconditional identity fact.
R3: If a sample has multiple supported taxon assignments, do not infer that the sample is a pure isolate.
R4: If quality controls indicate possible contamination, mark affected observations and downstream claims as qualified.
R5: If collection provenance is missing, label source claims unresolved; do not reconstruct missing facts.
R6: If methods or reference databases differ, preserve method-specific assignments and flag disagreements.
R7: A molecular detection does not by itself establish that an organism is viable.
R8: A source association does not by itself establish causation, pathogenicity or infection.
R9: An invalidated parent record triggers review of dependent claims without automatically invalidating independently supported evidence.

7. RECURSIVE PATHWAYS
P1 Provenance: sample -> collection -> source -> recorded context -> provenance confidence.
P2 Derivation: sample -> processing -> aliquot/isolate -> new observations -> qualified interpretation.
P3 Identity: observation -> method -> reference -> candidate classification -> validation boundary.
P4 Quality: sample -> control evidence -> issue detection -> affected results -> review.
P5 Knowledge update: new observation -> relation expansion -> boundary filter -> evidence comparison -> versioned update.
Recursion must be finite, cycle-aware and evidence-preserving.

8. TRI-MAPPING
Human concept | Ontology entity | Executable relation
"Where did it come from?" | SourceMaterial/CollectionEvent | derived_from/collected_during
"What is in it?" | MicrobialCommunity/Taxon | may_contain/assigns_taxon
"What was done to it?" | ProcessingEvent | consumes/produces
"Can we trust the result?" | QualityRecord/Evidence | assesses/supported_by
"Is it the same sample?" | SampleEntity/Identifier | identity and derivation constraints

9. CONFIDENCE AND UNCERTAINTY
For each claim c retain: evidence_ids, source_ids, method_id, timestamp, status, uncertainty.
confidence(c) = calibrated_assessment(evidence_quality, reproducibility, provenance, method_fit)
Do not invent numeric confidence scores without calibration data.
Distinguish absence of evidence from evidence of absence.
Retain alternative hypotheses when the evidence cannot distinguish them.

10. EXECUTION GUARANTEES
G1 No inferred identity without traceable supporting evidence.
G2 No sample-to-aliquot equivalence without an explicit identity policy.
G3 No unqualified inference of viability, pathogenicity, purity or clinical significance.
G4 All derived relationships are traceable to source records and rules.
G5 Contradictory observations are retained and flagged, not silently overwritten.
G6 Missing inputs yield 'unknown', not fabricated values.
G7 Ontology reasoning is conceptual; not a laboratory protocol or diagnostic determination.

11. AI SEARCH FLOW
flow "MicrobialSampleSearch" v1.0
  intent: resolve queries about microbial sample entities and their evidence
  normalise: synonyms, sample identifiers, taxon names, context, time
  retrieve: source records, derivation graph, observations, controls, reference versions
  expand: apply R1..R9 under B
  rank: provenance quality, method relevance, evidence specificity, uncertainty
  answer: classification + supporting evidence + limitations + linked entities
  abstain: when evidence is missing, conflicting or outside the declared boundary

12. UPDATE MODEL
On new evidence: append versioned observation; recompute affected inferred edges; preserve audit history.
On reclassification: retain prior taxonomic label and reference version; add superseding assertion.
On correction: record reason, actor/source and timestamp; never erase provenance silently.

13. CITATION
Title: Microbial Sample Ontology — Flow Algebra Companion
Authors: Mark Anthony Higgins; Sky — Recursive AI
Original page: https://kirethresearch.org/booksai-core/entity-graph/biological-entities/experimental-biological-entities/microbial-sample-ontology.html
Use: Open research companion; cite Kireth Research and the source page when reusing.