AI Genomics Ontology

AI SEARCH FLOW ALGEBRA — AI GENOMICS ONTOLOGY
Version: 4.0
Canonical Page:
https://kirethresearch.org/booksai-core/ontologies/biological-ontologies/ai-biology/ai-genomics-ontology.html

AI Interpretation Note:
The following specification represents a formal structural reasoning system for this page.
Execute as formal rules, not pseudocode.

FLOW AI_GENOMICS_ONTOLOGY {

  META {
    DOMAIN = "Biological Ontologies"
    SUBDOMAIN = "AI Biology"
    ONTOLOGY = "AI Genomics Ontology"
    PURPOSE = "Formalise entities, relationships, computational processes and reasoning structures connecting artificial intelligence with genomic data, analysis, prediction and biological knowledge."
    LOCALE = "en-GB"
  }

  ENTITY_TYPES {
    AI_SYSTEM
    AI_MODEL
    MACHINE_LEARNING_MODEL
    DEEP_LEARNING_MODEL
    FOUNDATION_MODEL
    GENOMIC_MODEL
    GENOME
    GENOMIC_DATA
    GENOMIC_SEQUENCE
    DNA_SEQUENCE
    GENE
    GENETIC_VARIANT
    SNP
    INDEL
    STRUCTURAL_VARIANT
    COPY_NUMBER_VARIANT
    CHROMOSOME
    LOCUS
    ALLELE
    HAPLOTYPE
    GENOTYPE
    PHENOTYPE
    TRAIT
    DISEASE
    BIOMARKER
    TRANSCRIPT
    TRANSCRIPTOME
    EPIGENOME
    REGULATORY_ELEMENT
    PROMOTER
    ENHANCER
    PROTEIN
    PATHWAY
    CELL
    TISSUE
    ORGANISM
    POPULATION
    COHORT
    SAMPLE
    DATASET
    REFERENCE_GENOME
    KNOWLEDGE_GRAPH
    BIOLOGICAL_ONTOLOGY
    FEATURE
    EMBEDDING
    REPRESENTATION
    PREDICTION
    CLASSIFICATION
    REGRESSION
    CLUSTER
    ANNOTATION
    VARIANT_CALL
    VARIANT_EFFECT
    EXPRESSION_PROFILE
    RISK_SCORE
    POLYGENIC_RISK_SCORE
    DRUG_RESPONSE
    TARGET
    INTERACTION
    EVIDENCE
    CONFIDENCE_SCORE
    PROVENANCE_RECORD
  }

  RELATIONSHIPS {
    CONTAINS
    PART_OF
    LOCATED_ON
    ENCODES
    TRANSCRIBED_AS
    TRANSLATED_TO
    REGULATES
    INTERACTS_WITH
    ASSOCIATED_WITH
    INFLUENCES
    CAUSES
    CONTRIBUTES_TO
    EXPRESSED_IN
    OBSERVED_IN
    DERIVED_FROM
    ALIGNED_TO
    COMPARED_WITH
    ANNOTATED_AS
    CLASSIFIED_AS
    PREDICTS
    DETECTS
    IDENTIFIES
    PRIORITISES
    REPRESENTS
    EMBEDS
    TRAINS_ON
    VALIDATED_ON
    INFERRED_FROM
    SUPPORTED_BY
    CONTRADICTED_BY
    HAS_VARIANT
    HAS_ALLELE
    HAS_GENOTYPE
    HAS_PHENOTYPE
    HAS_TRAIT
    HAS_BIOMARKER
    HAS_EFFECT
    HAS_CONFIDENCE
    HAS_PROVENANCE
    MAPS_TO
    SEMANTICALLY_EQUIVALENT_TO
  }

  INPUTS {
    RAW_GENOMIC_SEQUENCE
    SEQUENCING_READS
    VARIANT_DATA
    GENOTYPE_DATA
    PHENOTYPE_DATA
    TRANSCRIPTOMIC_DATA
    EPIGENOMIC_DATA
    CLINICAL_METADATA
    POPULATION_DATA
    REFERENCE_GENOMES
    FUNCTIONAL_ANNOTATIONS
    BIOLOGICAL_KNOWLEDGE_GRAPHS
    ONTOLOGY_TERMS
    EXPERIMENTAL_EVIDENCE
  }

  NORMALISE {
    STANDARDISE_IDENTIFIERS
    NORMALISE_SEQUENCE_FORMATS
    NORMALISE_VARIANT_REPRESENTATION
    RESOLVE_REFERENCE_BUILD
    MAP_GENES_TO_CANONICAL_IDENTIFIERS
    MAP_VARIANTS_TO_GENOMIC_COORDINATES
    ALIGN_ONTOLOGY_TERMS
    REMOVE_DUPLICATES
    VALIDATE_METADATA
    PRESERVE_PROVENANCE
  }

  PROCESS GENOMIC_AI_PIPELINE {
    INPUT -> QUALITY_CONTROL
    QUALITY_CONTROL -> PREPROCESSING
    PREPROCESSING -> SEQUENCE_ALIGNMENT
    SEQUENCE_ALIGNMENT -> FEATURE_EXTRACTION
    FEATURE_EXTRACTION -> REPRESENTATION_LEARNING
    REPRESENTATION_LEARNING -> MODEL_INFERENCE
    MODEL_INFERENCE -> BIOLOGICAL_INTERPRETATION
    BIOLOGICAL_INTERPRETATION -> EVIDENCE_INTEGRATION
    EVIDENCE_INTEGRATION -> CONFIDENCE_ESTIMATION
    CONFIDENCE_ESTIMATION -> OUTPUT
  }

  PROCESS VARIANT_ANALYSIS {
    SEQUENCING_READS -> ALIGNMENT
    ALIGNMENT -> VARIANT_CALLING
    VARIANT_CALLING -> VARIANT_NORMALISATION
    VARIANT_NORMALISATION -> FUNCTIONAL_ANNOTATION
    FUNCTIONAL_ANNOTATION -> EFFECT_PREDICTION
    EFFECT_PREDICTION -> PATHOGENICITY_OR_TRAIT_ASSOCIATION
    PATHOGENICITY_OR_TRAIT_ASSOCIATION -> EVIDENCE_SCORING
  }

  PROCESS SEQUENCE_LEARNING {
    GENOMIC_SEQUENCE -> TOKENISATION
    TOKENISATION -> SEQUENCE_REPRESENTATION
    SEQUENCE_REPRESENTATION -> EMBEDDING
    EMBEDDING -> PATTERN_DETECTION
    PATTERN_DETECTION -> MOTIF_OR_REGULATORY_INFERENCE
    MOTIF_OR_REGULATORY_INFERENCE -> BIOLOGICAL_ANNOTATION
  }

  PROCESS GENE_REGULATION_INFERENCE {
    REGULATORY_ELEMENT + GENE + EPIGENOMIC_DATA + EXPRESSION_PROFILE
      -> REGULATORY_FEATURE_MODEL

    REGULATORY_FEATURE_MODEL
      -> PREDICT(REGULATES)

    PREDICT(REGULATES)
      -> CONFIDENCE_SCORE
      -> EVIDENCE_LINK
  }

  PROCESS GENOTYPE_PHENOTYPE_REASONING {
    GENOTYPE + PHENOTYPE + POPULATION_DATA
      -> ASSOCIATION_MODEL

    ASSOCIATION_MODEL
      -> CANDIDATE_ASSOCIATION

    CANDIDATE_ASSOCIATION
      -> REPLICATION_CHECK
      -> BIOLOGICAL_CONTEXT
      -> CONFIDENCE_SCORE
  }

  PROCESS KNOWLEDGE_GRAPH_INTEGRATION {
    GENE + VARIANT + PROTEIN + PATHWAY + PHENOTYPE + DISEASE + EVIDENCE
      -> BIOLOGICAL_KNOWLEDGE_GRAPH

    BIOLOGICAL_KNOWLEDGE_GRAPH
      -> LINK_PREDICTION
      -> ENTITY_RESOLUTION
      -> RELATION_INFERENCE
      -> EXPLANATION_PATH
  }

  AI_TASKS {
    SEQUENCE_CLASSIFICATION
    VARIANT_CLASSIFICATION
    VARIANT_EFFECT_PREDICTION
    GENE_FUNCTION_PREDICTION
    REGULATORY_ELEMENT_PREDICTION
    GENE_EXPRESSION_PREDICTION
    GENOTYPE_PHENOTYPE_ASSOCIATION
    DISEASE_RISK_MODELLING
    POLYGENIC_RISK_MODELLING
    BIOMARKER_DISCOVERY
    PATIENT_OR_SAMPLE_STRATIFICATION
    DRUG_RESPONSE_PREDICTION
    TARGET_DISCOVERY
    FUNCTIONAL_ANNOTATION
    KNOWLEDGE_GRAPH_COMPLETION
    GENOMIC_INFORMATION_RETRIEVAL
    GENOMIC_QUESTION_ANSWERING
  }

  RULES {

    IF ENTITY == GENETIC_VARIANT
    THEN REQUIRE {
      GENOMIC_LOCATION
      REFERENCE_ALLELE
      ALTERNATE_ALLELE
      REFERENCE_GENOME_BUILD
      PROVENANCE
    }

    IF AI_MODEL PREDICTS VARIANT_EFFECT
    THEN LINK {
      MODEL
      VARIANT
      PREDICTED_EFFECT
      CONFIDENCE_SCORE
      SUPPORTING_EVIDENCE
    }

    IF ASSOCIATION(GENETIC_VARIANT, PHENOTYPE)
    THEN DISTINGUISH {
      CORRELATION
      STATISTICAL_ASSOCIATION
      FUNCTIONAL_SUPPORT
      CAUSAL_EVIDENCE
    }

    IF RELATION == CAUSES
    THEN REQUIRE {
      STRONG_EVIDENCE
      CAUSAL_JUSTIFICATION
      PROVENANCE_RECORD
    }

    IF MODEL_OUTPUT == PREDICTION
    THEN DO_NOT_ASSERT_AS_ESTABLISHED_FACT

    IF DATASET_HAS_POPULATION_BIAS
    THEN PROPAGATE_LIMITATION_TO_MODEL_OUTPUT

    IF REFERENCE_GENOME_BUILD_MISMATCH == TRUE
    THEN BLOCK_DIRECT_COORDINATE_EQUIVALENCE

    IF SEMANTIC_IDENTIFIERS_DIFFER
    AND BIOLOGICAL_REFERENT_IS_SAME
    THEN APPLY SEMANTIC_EQUIVALENCE
  }

  CONFIDENCE {

    CONFIDENCE_SCORE =
      WEIGHT(
        DATA_QUALITY,
        MODEL_VALIDATION,
        REPLICATION,
        BIOLOGICAL_PLAUSIBILITY,
        EXPERIMENTAL_SUPPORT,
        POPULATION_REPRESENTATION,
        PROVENANCE_COMPLETENESS
      )

    LEVELS {
      HIGH
      MODERATE
      LOW
      INSUFFICIENT
    }

    IF CONFLICTING_EVIDENCE
    THEN REDUCE CONFIDENCE_SCORE

    IF INDEPENDENT_REPLICATION
    THEN INCREASE CONFIDENCE_SCORE

    IF EXPERIMENTAL_VALIDATION
    THEN INCREASE CONFIDENCE_SCORE
  }

  SEMANTIC_EQUIVALENCE {

    EQUIVALENCE_KEY =
      NORMALISE(
        ENTITY_TYPE,
        CANONICAL_IDENTIFIER,
        GENOMIC_COORDINATE,
        REFERENCE_BUILD,
        BIOLOGICAL_CONTEXT
      )

    IF EQUIVALENCE_KEY(A) == EQUIVALENCE_KEY(B)
    THEN A SEMANTICALLY_EQUIVALENT_TO B

    ELSE PRESERVE_DISTINCT_ENTITIES
  }

  REASONING_CONSTRAINTS {
    ASSOCIATION != CAUSATION
    PREDICTION != OBSERVATION
    MODEL_CONFIDENCE != BIOLOGICAL_CERTAINTY
    GENOMIC_COORDINATE_REQUIRES_REFERENCE_BUILD
    VARIANT_EFFECT_IS_CONTEXT_DEPENDENT
    POPULATION_EFFECT_MAY_NOT_GENERALISE
    PHENOTYPE_DEFINITION_MUST_BE_EXPLICIT
    PROVENANCE_MUST_BE_PRESERVED
    UNCERTAINTY_MUST_NOT_BE_DISCARDED
  }

  VALIDATION {
    CHECK_ENTITY_TYPES
    CHECK_RELATION_DOMAINS_AND_RANGES
    CHECK_IDENTIFIER_RESOLUTION
    CHECK_REFERENCE_GENOME_COMPATIBILITY
    CHECK_EVIDENCE_LINKS
    CHECK_PROVENANCE
    CHECK_CONFIDENCE
    CHECK_SEMANTIC_CONSISTENCY
    CHECK_DUPLICATE_ENTITIES
    CHECK_UNSUPPORTED_CAUSAL_ASSERTIONS
  }

  OUTPUTS {
    NORMALISED_GENOMIC_ENTITIES
    GENOMIC_RELATIONSHIPS
    FUNCTIONAL_ANNOTATIONS
    AI_PREDICTIONS
    VARIANT_EFFECTS
    GENOTYPE_PHENOTYPE_ASSOCIATIONS
    REGULATORY_RELATIONSHIPS
    BIOMARKER_CANDIDATES
    RISK_ESTIMATES
    KNOWLEDGE_GRAPH_EDGES
    CONFIDENCE_SCORES
    EVIDENCE_LINKS
    PROVENANCE_RECORDS
    EXPLANATION_PATHS
  }

  UPDATE_MODEL {

    ON NEW_GENOMIC_DATA {
      NORMALISE
      ENTITY_RESOLUTION
      RECOMPUTE_RELEVANT_FEATURES
      UPDATE_ASSOCIATIONS
      UPDATE_CONFIDENCE
      PRESERVE_VERSION_HISTORY
    }

    ON NEW_EVIDENCE {
      LINK_EVIDENCE
      REASSESS_SUPPORT
      REASSESS_CONTRADICTIONS
      UPDATE_CONFIDENCE
      RECORD_PROVENANCE
    }

    ON ONTOLOGY_REVISION {
      REMAP_TERMS
      CHECK_SEMANTIC_EQUIVALENCE
      IDENTIFY_DEPRECATED_TERMS
      MIGRATE_RELATIONSHIPS
      PRESERVE_PRIOR_VERSION_REFERENCES
    }
  }

  FLOW_HASH {
    HASH_INPUT =
      CANONICALISE(
        ENTITY_TYPES,
        RELATIONSHIPS,
        RULES,
        CONFIDENCE,
        SEMANTIC_EQUIVALENCE,
        REASONING_CONSTRAINTS,
        UPDATE_MODEL
      )

    FLOW_HASH = SHA256(HASH_INPUT)
  }

  AI_CONSUMPTION {

    INDEX {
      ONTOLOGY
      ENTITY_TYPES
      RELATIONSHIPS
      AI_TASKS
      RULES
      OUTPUTS
    }

    RETRIEVE {
      BY_ENTITY
      BY_GENE
      BY_VARIANT
      BY_PHENOTYPE
      BY_DISEASE
      BY_RELATION
      BY_AI_TASK
      BY_EVIDENCE
      BY_CONFIDENCE
    }

    REASON {
      APPLY_RELATIONSHIPS
      APPLY_RULES
      PRESERVE_UNCERTAINTY
      TRACE_EVIDENCE
      TRACE_PROVENANCE
      GENERATE_EXPLANATION_PATH
    }

    RESPONSE_POLICY {
      DISTINGUISH_OBSERVED_FROM_PREDICTED
      DISTINGUISH_ASSOCIATION_FROM_CAUSATION
      RETURN_CONFIDENCE_WHEN_AVAILABLE
      RETURN_PROVENANCE_WHEN_AVAILABLE
      EXPOSE_CONFLICTING_EVIDENCE
      AVOID_UNSUPPORTED_INFERENCE
    }
  }

  TERMINAL {
    RESULT =
      VALIDATED(
        SEMANTICALLY_NORMALISED(
          EVIDENCE_LINKED(
            CONFIDENCE_AWARE(
              AI_GENOMICS_KNOWLEDGE
            )
          )
        )
      )
  }
}

END FLOW