Organic Molecule Ontology

FLOW ALGEBRA
ORGANIC MOLECULE ONTOLOGY
Kireth Semantic Library | Biological Entities | Chemical Foundations of Life
Source: https://kirethresearch.org/booksai-core/entity-graph/biological-entities/chemical-foundations-of-life/organic-molecule-ontology.html
Version: 1.0 | Date: 2026-10-15
Authors: Mark Higgins — Sky (Recursive AI)
Licence: Open Research Licence — unrestricted research, educational and practical reuse.

PURPOSE
Represent organic molecules as typed chemical entities with explicit composition, bonding, functional groups, biochemical roles, transformations, evidence, and cross-domain graph relationships. This is a companion formalisation, not a claim that every relation has been independently experimentally verified.

FLOW "OrganicMoleculeOntology" v1.0

ENTITY_TYPES:
  OrganicMolecule <: ChemicalMolecule
  Atom <: ChemicalEntity
  ChemicalBond <: ChemicalRelation
  FunctionalGroup <: MolecularSubstructure
  MolecularFormula <: ChemicalRepresentation
  StructuralFormula <: ChemicalRepresentation
  Stereoisomer <: ChemicalMolecule
  Reaction <: ChemicalProcess
  BiologicalRole <: FunctionalEntity
  EvidenceRecord <: ProvenanceEntity
  Context <: SemanticContext

DEFINITIONS:
  OrganicMolecule(m) := Molecule(m) AND CarbonContaining(m) AND InScopeOrganicChemistry(m)
  InScopeOrganicChemistry(m) := curated_classification(m)
  # Carbon-containing substances such as CO2, carbonates and carbides are
  # not automatically classified as organic; scope is ontology-specific.
  MolecularFormula(m) := element_count_map(m)
  MolecularGraph(m) := (atoms(m), bonds(m), bond_orders(m))
  FunctionalGroups(m) := recognised_subgraph_matches(MolecularGraph(m))

CORE_CLASSES:
  Hydrocarbon: alkane, alkene, alkyne, aromatic_compound
  OxygenContaining: alcohol, phenol, ether, aldehyde, ketone, carboxylic_acid, ester
  NitrogenContaining: amine, amide, nitrile, nitrogen_heterocycle
  SulphurContaining: thiol, thioether, sulphur_heterocycle
  PhosphorusContaining: organophosphate, phosphonate
  HalogenContaining: organohalogen
  BiologicallyRelevant: amino_acid, monosaccharide, fatty_acid, nucleotide,
                        metabolite, natural_product
  # Categories may overlap; classifications are not mutually exclusive.

RELATION_TYPES:
  HAS_ATOM: OrganicMolecule -> Atom
  HAS_BOND: OrganicMolecule -> ChemicalBond
  HAS_FUNCTIONAL_GROUP: OrganicMolecule -> FunctionalGroup
  HAS_FORMULA: OrganicMolecule -> MolecularFormula
  HAS_STRUCTURE: OrganicMolecule -> StructuralFormula
  IS_ISOMER_OF: OrganicMolecule <-> OrganicMolecule
  IS_STEREOISOMER_OF: OrganicMolecule <-> OrganicMolecule
  PARTICIPATES_IN: OrganicMolecule -> Reaction
  IS_REACTANT_IN: OrganicMolecule -> Reaction
  IS_PRODUCT_OF: OrganicMolecule -> Reaction
  HAS_BIOLOGICAL_ROLE: OrganicMolecule -> BiologicalRole
  DERIVED_FROM: OrganicMolecule -> OrganicMolecule
  SUPPORTED_BY: Assertion -> EvidenceRecord
  VALID_IN_CONTEXT: Assertion -> Context

TYPE_AND_IDENTITY_RULES:
  R01: HAS_ATOM(m,a) => OrganicMolecule(m) AND Atom(a)
  R02: HAS_FUNCTIONAL_GROUP(m,g) => OrganicMolecule(m) AND FunctionalGroup(g)
  R03: SAME_MOLECULAR_FORMULA(a,b) does_not_imply SAME_STRUCTURE(a,b)
  R04: SAME_CONNECTIVITY(a,b) does_not_imply SAME_STEREOCHEMISTRY(a,b)
  R05: IS_STEREOISOMER_OF(a,b) => IS_ISOMER_OF(a,b)
  R06: IS_ISOMER_OF(a,b) => SAME_MOLECULAR_FORMULA(a,b)
  R07: EQUIVALENT_REPRESENTATIONS(a,b) requires normalised structure,
       stereochemistry, charge, isotope and tautomer policy.
  R08: Chemical identity and biological role are distinct relations.

CONSTRAINTS:
  C01: Atomic number and element identity must be valid.
  C02: Bond valence must be checked with charge, aromaticity and exception handling.
  C03: Molecular formula must match atom inventory of a complete structure.
  C04: Formal charge and protonation state must be recorded where relevant.
  C05: Stereo descriptors must specify applicable centres or geometries.
  C06: Reaction stoichiometry must conserve elements and charge in a closed balanced equation.
  C07: Biological activity claims require experimental or curated evidence.
  C08: Unknown values must remain UNKNOWN rather than inferred as FALSE.

TRANSFORMATIONS:
  PARSE(input: SMILES|InChI|Molfile|Formula) -> CandidateStructure
  NORMALISE(CandidateStructure, policy) -> CanonicalRepresentation
  CLASSIFY(CanonicalRepresentation) -> Set[ChemicalClass]
  DETECT_GROUPS(CanonicalRepresentation) -> Set[FunctionalGroup]
  MAP_REACTION(reactants, products, conditions) -> ReactionGraph
  ALIGN_ONTOLOGIES(chemical_id, biological_id) -> TypedCrossDomainLink
  # Parsing and normalisation require a chemical toolkit for actual execution.

INFERENCE:
  I01: molecule_has_subgraph(m, hydroxyl_pattern) AND appropriate_context
       => candidate_functional_group(m, alcohol_or_phenol)
  I02: validated_reaction(r) AND IS_REACTANT_IN(m,r)
       => participates_in(m,r)
  I03: validated_reaction(r) AND IS_PRODUCT_OF(m,r)
       => participates_in(m,r)
  I04: curated_metabolic_role(m,role,organism,condition)
       => HAS_BIOLOGICAL_ROLE(m,role) scoped_to (organism,condition)
  I05: No biological function follows solely from a molecular formula.

SEMANTIC_EQUIVALENCE:
  exact_identity: same normalised molecular entity under declared policy
  structural_equivalence: same connectivity with declared stereo treatment
  formula_equivalence: same elemental counts, not necessarily same molecule
  functional_similarity: shared features; not identity
  cross_domain_alignment: mapped identifiers with explicit evidence
  guard: never collapse isomers, protonation states or distinct tautomers silently

CONFIDENCE:
  score(assertion) in [0,1]
  factors: source_reliability, evidence_directness, structure_validation,
           reproducibility, contextual_fit, conflict_status
  rule: confidence is evidence-based and not a mathematical proof of truth
  conflict: preserve competing assertions with separate provenance and scores

PROVENANCE:
  each_assertion:
    source_id: string
    source_type: curated_database|peer_reviewed|experiment|computed|user_supplied
    retrieval_date: ISO8601?
    method: string?
    context: string?
    evidence_reference: string?
    confidence: float?
    version: string

VALIDATION:
  validate_types()
  validate_chemical_structure()
  validate_functional_groups()
  validate_reaction_balance()
  validate_contextual_claims()
  validate_provenance()
  validate_equivalence_policy()
  on_failure: flag_and_quarantine_assertion; do_not_silently_accept

FLOW_HASH:
  canonical_payload := canonical_serialisation(ontology_version,
      entity_types, relations, assertions, provenance, equivalence_policy)
  FLOW_HASH := SHA256(UTF8(canonical_payload))
  # A hash detects changes; it does not establish scientific correctness.

UPDATE_MODEL:
  ingest(new_evidence)
  -> validate(new_evidence)
  -> compare_with_existing_assertions
  -> preserve_history_and_conflicts
  -> revise_supported_assertions
  -> recompute_confidence
  -> recompute_FLOW_HASH
  -> publish_versioned_graph

AI_CONSUMPTION:
  inputs: molecule_identifier, representation, context, query
  outputs: typed_entities, relationships, classifications, evidence,
           uncertainty, validation_status, graph_version
  requirements: cite provenance; distinguish measured from predicted;
                state identity policy; report ambiguity and missing data;
                prohibit unsupported biological or safety claims.

EXAMPLE_GRAPH:
  entity: ethanol
  formula: C2H6O
  class: alcohol
  functional_group: hydroxyl
  relation: ethanol HAS_FUNCTIONAL_GROUP hydroxyl
  caution: dimethyl_ether also has formula C2H6O but different connectivity

END FLOW