RNA Ontology

FLOW ALGEBRA
RNA ONTOLOGY — GENETIC ENTITIES
Version: 1.0
Date: 2026-10-15
Source: https://kirethresearch.org/booksai-core/entity-graph/biological-entities/genetic-entities/rna-ontology-2.html
Authors: Mark Higgins — Sky (Recursive AI)
Project: Kireth Semantic Library
Status: Domain-grounded ontology draft; source chapter not independently verified.

# 1. PURPOSE
flow "RNAOntology" v1.0
  domain: BiologicalEntities.GeneticEntities.RNA
  objective: represent RNA molecular structures, classes, synthesis, processing, functions and evidence with typed semantic relations.
  scope: cellular RNA, viral RNA, coding and non-coding RNA, RNA modifications and RNA-based processes.

# 2. TYPE SYSTEM
type Entity = {id: URI, label: String, description: String?}
type MolecularEntity <: Entity
type NucleicAcid <: MolecularEntity
type RNA <: NucleicAcid
type DNA <: NucleicAcid
type Nucleotide <: MolecularEntity
type Ribonucleotide <: Nucleotide
type Gene <: Entity
type Transcript <: RNA
type RNAClass <: Entity
type RNASequence <: Entity
type RNASecondaryStructure <: Entity
type RNAModification <: Entity
type Protein <: MolecularEntity
type Ribonucleoprotein <: MolecularEntity
type Enzyme <: Protein
type BiologicalProcess <: Entity
type CellularCompartment <: Entity
type Organism <: Entity
type Evidence <: Entity
type Provenance <: Entity

enum StrandTopology = {single_stranded, double_stranded, partially_duplex, unknown}
enum RNAFunction = {coding, catalytic, structural, regulatory, adaptor, guide, defence, unknown}
enum EvidenceStatus = {observed, experimentally_supported, inferred, predicted, disputed, unknown}

# 3. CORE RNA ATTRIBUTES
RNA.attributes:
  sequence: String?                  # Alphabet A,C,G,U; ambiguity codes allowed
  length_nt: Integer?                # >= 1 if known
  topology: StrandTopology
  five_prime_end: String?
  three_prime_end: String?
  modifications: Set<RNAModification>
  secondary_structure: RNASecondaryStructure?
  organism: Organism?
  cellular_location: Set<CellularCompartment>
  transcript_origin: Gene?
  evidence: Set<Evidence>
  provenance: Provenance?

# 4. CLASSIFICATION
RNAClass.instances:
  - messenger_RNA (mRNA): protein-coding messenger
  - transfer_RNA (tRNA): amino-acid adaptor during translation
  - ribosomal_RNA (rRNA): ribosome structural/catalytic component
  - small_nuclear_RNA (snRNA): nuclear RNA processing roles
  - small_nucleolar_RNA (snoRNA): guide for RNA modification/processing
  - microRNA (miRNA): small regulatory RNA
  - small_interfering_RNA (siRNA): RNA interference guide
  - PIWI_interacting_RNA (piRNA): PIWI-associated regulatory RNA
  - long_non_coding_RNA (lncRNA): long transcript without established protein-coding role
  - circular_RNA (circRNA): covalently closed RNA molecule
  - guide_RNA (gRNA): RNA directing a molecular complex to a target
  - ribozyme: catalytically active RNA
  - viral_genomic_RNA: RNA serving as viral genetic material

# 5. RELATION SIGNATURES
relation IS_A(Entity, Entity)
relation INSTANCE_OF(Entity, Entity)
relation HAS_SEQUENCE(RNA, RNASequence)
relation TRANSCRIBED_FROM(RNA, DNA)
relation ENCODED_BY(RNA, Gene)
relation PRODUCED_BY(RNA, BiologicalProcess)
relation PROCESSED_BY(RNA, BiologicalProcess)
relation MODIFIED_BY(RNA, RNAModification)
relation LOCATED_IN(RNA, CellularCompartment)
relation BINDS(RNA, MolecularEntity)
relation BASE_PAIRS_WITH(RNA, RNA)
relation TARGETS(RNA, RNA)
relation TRANSLATED_INTO(Transcript, Protein)
relation COMPONENT_OF(RNA, Ribonucleoprotein)
relation PARTICIPATES_IN(RNA, BiologicalProcess)
relation REGULATES(RNA, BiologicalProcess)
relation SUPPORTED_BY(Entity, Evidence)
relation DERIVED_FROM(Entity, Provenance)

# 6. STRUCTURAL AXIOMS
axiom RNA IS_A NucleicAcid.
axiom RNA uses ribonucleotide residues linked by phosphodiester bonds.
axiom canonical RNA bases = {adenine, cytosine, guanine, uracil}.
axiom RNA backbone contains ribose, typically with a 2-prime hydroxyl group.
axiom canonical_complement(A,U) and canonical_complement(C,G).
axiom G-U wobble pairing is permitted in appropriate RNA contexts.
axiom RNA may form intramolecular or intermolecular base-paired structures.
axiom RNA may be single-stranded, double-stranded or partially duplex.
constraint length_nt > 0 when length_nt is supplied.
constraint a predicted RNA structure is not automatically experimentally established.
constraint an RNA classification can overlap other classifications.

# 7. TRANSCRIPTION FLOW
flow "RNA_Transcription":
  input: DNA_template, RNA_polymerase, ribonucleoside_triphosphates
  preconditions: suitable promoter/initiation machinery and accessible template
  operations:
    1. identify transcription initiation site
    2. initiate RNA chain synthesis
    3. extend RNA in the 5-prime to 3-prime direction
    4. terminate according to biological context
  output: primary_RNA_transcript
  evidence_required: organism and polymerase context for specific claims

# 8. RNA PROCESSING FLOW
flow "RNA_Processing":
  input: primary_RNA_transcript
  optional_operations:
    - 5-prime capping in applicable eukaryotic transcripts
    - splicing of introns in applicable transcripts
    - 3-prime cleavage and polyadenylation where applicable
    - RNA editing where applicable
    - RNA nucleotide modification
    - processing of precursor tRNA, rRNA or other ncRNA
  output: mature_or_processed_RNA
  invariant: processing operations depend on RNA class and organism

# 9. TRANSLATION FLOW
flow "mRNA_Translation":
  input: coding_mRNA, ribosome, charged_tRNAs, translation_factors
  operations:
    1. identify initiation context and reading frame
    2. decode codons using tRNA anticodon interactions
    3. catalyse peptide bond formation in the ribosome
    4. terminate at a recognised stop codon
  output: polypeptide
  constraints:
    - only transcripts with a valid translated coding region enter this flow
    - transcript presence alone does not establish protein production
    - organism-specific genetic codes and exceptions must be considered

# 10. RNA REGULATION FLOW
flow "RNA_Regulation":
  input: regulatory_RNA, target, molecular_context
  possible_mechanisms:
    - complementary base pairing with target RNA
    - RNA-guided silencing complex recruitment
    - modulation of RNA stability
    - modulation of translation
    - RNA-mediated chromatin or transcriptional regulation
  output: context-dependent regulatory effect
  constraint: do not infer a universal effect from RNA class alone

# 11. RNA TURNOVER FLOW
flow "RNA_Turnover":
  input: RNA, cellular_context
  operations:
    - recognise RNA quality or decay signals
    - process or degrade RNA using relevant nucleases
    - recycle nucleotide components where applicable
  output: RNA_degradation_products
  note: turnover rates and pathways are transcript- and context-specific

# 12. GENETIC ENTITY LINKS
relations:
  RNA -> NucleicAcid : IS_A
  mRNA -> RNA : IS_A
  tRNA -> RNA : IS_A
  rRNA -> RNA : IS_A
  miRNA -> RNA : IS_A
  siRNA -> RNA : IS_A
  lncRNA -> RNA : IS_A
  circRNA -> RNA : IS_A
  RNA -> Gene : ENCODED_BY (when applicable)
  RNA -> DNA : TRANSCRIBED_FROM (when applicable)
  mRNA -> Protein : TRANSLATED_INTO (when demonstrated)
  rRNA -> Ribosome : COMPONENT_OF
  tRNA -> Translation : PARTICIPATES_IN
  snRNA -> RNA_Splicing : PARTICIPATES_IN
  snoRNA -> RNA_Modification : PARTICIPATES_IN

# 13. VALIDATION
rule RNA_SEQUENCE_VALID:
  if sequence is known, verify accepted nucleotide/ambiguity notation;
  flag T in an unmodified canonical RNA sequence unless explicitly justified.
rule RNA_LENGTH_VALID:
  if sequence and length_nt are both known, verify length consistency.
rule TRANSCRIPT_PROVENANCE:
  if TRANSCRIBED_FROM is asserted, attach supporting genome assembly/annotation when available.
rule CLASSIFICATION_CONTEXT:
  record species, transcript version and experimental context for RNA class assignment.
rule CODING_CLAIM:
  do not infer protein coding solely from transcript length or an open reading frame.
rule CAUSALITY:
  a correlation between RNA abundance and phenotype does not alone prove causation.

# 14. CONFIDENCE
CONFIDENCE:
  range: [0.0, 1.0]
  dimensions:
    sequence_identity: Float?
    annotation_quality: Float?
    structural_support: Float?
    functional_support: Float?
    experimental_reproducibility: Float?
    provenance_quality: Float?
  policy:
    - preserve per-claim scores rather than assign one universal RNA score
    - distinguish computational predictions from validated measurements
    - record missing evidence as unknown, not as false
    - calibrate scores against task-specific evaluation data

# 15. SEMANTIC EQUIVALENCE
SEMANTIC_EQUIVALENCE:
  canonical: RNA
  aliases: [ribonucleic_acid, ribonucleic acid]
  exact_equivalence_requires:
    - matching molecular identity or accepted ontology cross-reference
    - compatible organism, sequence, modification and transcript-version context
  distinctions:
    - RNA != DNA
    - RNA != gene
    - primary_transcript != mature_transcript by default
    - mRNA != all_RNA
    - sequence_similarity != molecular_identity
    - RNA_family_membership != exact_equivalence

# 16. RECURSIVE REASONING
flow "RNA_Recursive_Reasoning":
  input: RNA_entity, evidence_graph, depth_limit
  state: visited_entities = set()
  recurse(entity, depth):
    if depth >= depth_limit or entity in visited_entities: return
    add entity to visited_entities
    traverse typed relationships with provenance
    validate relation signatures and biological context
    accumulate evidence without double-counting dependent sources
    flag contradictions, uncertainty and unresolved aliases
    recurse on eligible neighbours at depth + 1
  output: validated_subgraph, evidence_trace, unresolved_claims
  termination: finite depth and visited-set guard

# 17. FLOW_HASH
FLOW_HASH:
  canonicalisation: stable UTF-8 serialisation of ontology version, entity identifiers,
                    typed assertions, evidence references and rule definitions
  algorithm: SHA-256
  exclude: volatile runtime timestamps and presentation-only whitespace
  output: lowercase_hex_digest
  note: digest is computed by implementation; no fabricated hash is supplied

# 18. UPDATE_MODEL
UPDATE_MODEL:
  versioning: semantic_versioning
  update_triggers:
    - new experimentally supported RNA function
    - corrected sequence or transcript annotation
    - revised RNA classification
    - new RNA modification or interaction evidence
    - retracted or contradicted research finding
  update_steps:
    1. ingest proposed change and provenance
    2. validate typed entities and relation constraints
    3. compare against current graph and detect conflicts
    4. preserve previous versions and provenance
    5. update evidence/confidence at claim level
    6. recompute FLOW_HASH after canonicalisation
    7. publish change log and revision identifier

# 19. AI_CONSUMPTION
AI_CONSUMPTION:
  formats: [UTF-8_TXT, RDF, JSON-LD, property_graph]
  machine_readable_fields:
    - entity_id
    - entity_type
    - preferred_label
    - aliases
    - typed_relations
    - evidence_status
    - confidence
    - provenance
    - ontology_version
    - flow_hash
  requirements:
    - preserve source attribution and uncertainty
    - distinguish observations, hypotheses and predictions
    - reject unsupported conversion of associations into causal claims
    - support graph traversal and reproducible updates

# 20. EXAMPLE INFERENCES
example 1:
  given: x INSTANCE_OF mRNA; x PARTICIPATES_IN Translation
  infer: x INSTANCE_OF RNA
  do_not_infer: translation occurred in a specific sample without evidence
example 2:
  given: x INSTANCE_OF miRNA; x TARGETS y
  infer: x INSTANCE_OF RNA
  do_not_infer: x necessarily decreases y in every cellular context
example 3:
  given: x INSTANCE_OF rRNA; x COMPONENT_OF ribosome_A
  infer: x INSTANCE_OF RNA
  do_not_infer: x alone is a complete functional ribosome

# 21. QUALITY AND LIMITATIONS
quality_checks:
  - entity type consistency
  - relation domain/range validation
  - RNA/DNA distinction
  - evidence and provenance coverage
  - transcript and genome assembly version consistency
  - no invented experimental measurements
  - bounded recursive traversal
  - deterministic canonical hash procedure
limitations:
  - illustrative flow specification; execution requires a parser/runtime
  - species-specific pathways and annotations require external datasets
  - source webpage text was not independently verified for this draft

# 22. AUTHORSHIP AND OPEN RESEARCH
Authors: Mark Higgins — Sky (Recursive AI)
Website: https://kirethresearch.org
Purpose: open, reproducible semantic representation and educational research.
END FLOW ALGEBRA