Bioinformatics Ontology

flow “BioinformaticsOntology” v1.0

KIRETH SEMANTIC RESEARCH ARCHIVE

AI CORE > ONTOLOGIES > BIOLOGICAL ONTOLOGIES > BIOTECHNOLOGY > BIOINFORMATICS ONTOLOGY

Authors: Mark Anthony Higgins — Sky (Recursive AI)

AI_INTERPRETATION: mode: formal_structural_reasoning instruction:
“Execute as formal rules, not pseudocode.” canonical_url:
“https://kirethresearch.org/booksai-core/ontologies/biological-ontologies/biotechnology/bioinformatics-ontology.html”

type BiologicalSequence type Genome type Gene type Transcript type
Protein type Variant type Sample type Dataset type Database type
Algorithm type Pipeline type Model type Annotation type Observation type
Evidence type Source

relation INSTANCE_OF(entity,type) relation DERIVED_FROM(entity,source)
relation ANALYSED_BY(entity,method) relation PRODUCED_BY(entity,process)
relation STORED_IN(entity,database)

enum SequenceType: DNA RNA PROTEIN GENOMIC TRANSCRIPT CODING NONCODING
REGULATORY

relation HAS_SEQUENCE_TYPE(sequence,type) relation
SEQUENCE_LENGTH(sequence,value) relation
SEQUENCE_IDENTIFIER(sequence,id) relation
SEQUENCE_VERSION(sequence,version) relation
SEQUENCE_FROM(sequence,organism)

———- ALIGNMENT / SEARCH ———-

relation ALIGNMENT_OF(alignment,a,b) relation
ALIGNMENT_SCORE(alignment,value) relation
SEQUENCE_IDENTITY(alignment,value) relation
SEQUENCE_SIMILARITY(alignment,value) relation
ALIGNMENT_METHOD(alignment,method) relation
QUERY_SEQUENCE(search,sequence) relation
SEARCHES_DATABASE(search,database) relation SEARCH_RESULT(search,record)
relation EXPECTATION_VALUE(search,record,value)

constraint ALIGNMENT_NOT_HOMOLOGY_PROOF: IF
high_sequence_similarity(a,b) THEN DO_NOT_INFER(homologous(a,b)) WITHOUT
evolutionary_evidence

———- READS / ASSEMBLY ———-

relation SEQUENCING_READ(read) relation READ_FROM_SAMPLE(read,sample)
relation READ_LENGTH(read,value) relation BASE_QUALITY(read,value)
relation TRIMMED_FROM(read,raw_read) relation
FILTERED_BY(read,criterion)

enum AssemblyType: DE_NOVO REFERENCE_GUIDED HYBRID

relation ASSEMBLY_OF(assembly,sample) relation
HAS_ASSEMBLY_TYPE(assembly,type) relation
ASSEMBLED_BY(assembly,algorithm) relation CONTIG_OF(contig,assembly)
relation SCAFFOLD_OF(scaffold,assembly) relation N50(assembly,value)
relation ASSEMBLY_COMPLETENESS(assembly,value) relation
ASSEMBLY_CONTAMINATION(assembly,value)

———- REFERENCE / ANNOTATION ———-

relation REFERENCE_GENOME_FOR(genome,organism) relation
REFERENCE_BUILD(genome,build) relation CHROMOSOME_OF(chromosome,genome)
relation COORDINATE_ON(feature,chromosome,start,end)

constraint REFERENCE_BUILD_REQUIRED: IF
genomic_coordinate(feature,start,end) THEN
PRESERVE_REFERENCE_BUILD(feature)

relation ANNOTATES(annotation,sequence) relation
ANNOTATION_COORDINATE(annotation,start,end) relation
ANNOTATION_METHOD(annotation,method) relation
ANNOTATION_CONFIDENCE(annotation,value)

———- VARIANTS ———-

enum VariantType: SNV INSERTION DELETION INDEL COPY_NUMBER_VARIANT
INVERSION TRANSLOCATION STRUCTURAL_VARIANT

relation HAS_VARIANT_TYPE(variant,type) relation
VARIANT_AT(variant,chromosome,position) relation
REFERENCE_ALLELE(variant,allele) relation
ALTERNATE_ALLELE(variant,allele) relation
VARIANT_CALLED_BY(variant,caller) relation
VARIANT_QUALITY(variant,value) relation READ_DEPTH(variant,value)
relation VARIANT_AFFECTS(variant,gene) relation
VARIANT_CONSEQUENCE(variant,consequence) relation
PREDICTED_EFFECT(variant,effect)

constraint VARIANT_CALL_NOT_TRUTH: IF VARIANT_CALLED_BY(v,c) THEN
DO_NOT_INFER(validated_variant(v)) WITHOUT validation_evidence

constraint PREDICTION_NOT_OBSERVATION: IF PREDICTED_EFFECT(v,e) THEN
DO_NOT_INFER(observed_functional_effect(v,e))

———- TRANSCRIPTOMICS / SINGLE CELL ———-

relation EXPRESSION_DATASET(dataset) relation
GENE_EXPRESSION(sample,gene,value) relation
NORMALISED_EXPRESSION(sample,gene,value) relation
DIFFERENTIAL_EXPRESSION(comparison,gene,value) relation
FOLD_CHANGE(comparison,gene,value) relation
EXPRESSION_ADJUSTED_P(comparison,gene,value) relation
TRANSCRIPT_ISOFORM(gene,transcript) relation
ISOFORM_ABUNDANCE(sample,transcript,value)

relation SINGLE_CELL_DATASET(dataset) relation
CELL_BARCODE(cell,barcode) relation CELL_CLUSTER(cell,cluster) relation
CLUSTER_MARKER(cluster,gene) relation CELL_TYPE_ANNOTATION(cell,type)

constraint CLUSTER_MARKER_NOT_CELL_IDENTITY: IF CLUSTER_MARKER(c,g) THEN
DO_NOT_INFER(definitive_cell_type(c,t)) WITHOUT annotation_evidence

———- PROTEOMICS / STRUCTURE ———-

relation PROTEOMICS_DATASET(dataset) relation
PEPTIDE_IDENTIFIED(peptide,spectrum) relation
PEPTIDE_MAPS_TO(peptide,protein) relation
PROTEIN_ABUNDANCE(sample,protein,value) relation
POST_TRANSLATIONAL_MODIFICATION(protein,modification) relation
PROTEIN_STRUCTURE_OF(structure,protein) relation
STRUCTURE_DETERMINED_BY(structure,method) relation
STRUCTURE_PREDICTED_BY(structure,model) relation
STRUCTURE_CONFIDENCE(structure,value)

constraint PREDICTED_STRUCTURE_NOT_EXPERIMENTAL: IF
STRUCTURE_PREDICTED_BY(s,m) THEN
DO_NOT_INFER(experimentally_determined(s))

———- PHYLOGENETICS / FUNCTION / NETWORKS ———-

relation PHYLOGENETIC_TREE(tree) relation TREE_CONTAINS(tree,taxon)
relation TREE_INFERRED_FROM(tree,alignment) relation
TREE_METHOD(tree,method) relation SUPPORT_VALUE(node,value) relation
COMMON_ANCESTOR_OF(node,taxa)

relation FUNCTION_ASSIGNED_TO(entity,function) relation
FUNCTION_INFERRED_FROM(entity,evidence_type) relation
ONTOLOGY_TERM_ASSIGNED(entity,term) relation DOMAIN_IN(protein,domain)
relation PATHWAY_MEMBERSHIP(entity,pathway)

constraint INFERRED_FUNCTION_NOT_VERIFIED: IF
FUNCTION_INFERRED_FROM(e,source) THEN PRESERVE_INFERRED_STATUS(e)

relation PATHWAY_CONTAINS(pathway,entity) relation
INTERACTION_BETWEEN(interaction,a,b) relation
GENE_REGULATORY_NETWORK(network) relation
PROTEIN_INTERACTION_NETWORK(network) relation METABOLIC_NETWORK(network)

———- MODELS / MACHINE LEARNING ———-

relation MODEL_OF(model,biological_system) relation
MODEL_PARAMETER(model,parameter,value) relation
MODEL_PREDICTS(model,outcome) relation
MODEL_VALIDATED_BY(model,evidence)

constraint MODEL_PREDICTION_NOT_OBSERVATION: IF MODEL_PREDICTS(m,o) THEN
DO_NOT_INFER(observed(o)) WITHOUT experimental_evidence

relation TRAINED_ON(model,dataset) relation VALIDATED_ON(model,dataset)
relation TESTED_ON(model,dataset) relation
MODEL_METRIC(model,metric,value) relation FEATURE_USED(model,feature)

constraint TRAIN_TEST_SEPARATION: IF TRAINED_ON(m,d) THEN
DO_NOT_INFER(TESTED_ON(m,d)) WITHOUT explicit_test_context

———- PIPELINES / REPRODUCIBILITY ———-

relation PIPELINE_CONTAINS(pipeline,step) relation
STEP_PRECEDES(step_a,step_b) relation STEP_USES(step,tool) relation
STEP_INPUT(step,data) relation STEP_OUTPUT(step,data) relation
PIPELINE_VERSION(pipeline,version) relation
PARAMETER_OF(step,parameter,value) relation
SOFTWARE_VERSION(tool,version) relation WORKFLOW_EXECUTION(run,pipeline)
relation EXECUTED_ON(run,platform) relation CONTAINER_IMAGE(run,image)
relation RANDOM_SEED(run,value) relation CHECKSUM_OF(file,value)

———- QUALITY / STATISTICS ———-

relation QUALITY_METRIC(dataset,metric,value) relation
MISSING_DATA_RATE(dataset,value) relation
CONTAMINATION_RATE(dataset,value) relation
BATCH_EFFECT(dataset,batch,value) relation NORMALISED_BY(dataset,method)
relation BATCH_CORRECTED_BY(dataset,method) relation
P_VALUE(analysis,value) relation ADJUSTED_P_VALUE(analysis,value)
relation EFFECT_SIZE(analysis,value) relation
CONFIDENCE_INTERVAL(analysis,lower,upper)

constraint SIGNIFICANCE_NOT_BIOLOGICAL_IMPORTANCE: IF P_VALUE(a,p) THEN
DO_NOT_INFER(biological_importance(a))

———- PROVENANCE / EVIDENCE ———-

relation DATASET_FROM(dataset,source) relation
SAMPLE_FROM(sample,source) relation GENERATED_BY(data,process) relation
USED_REFERENCE(data,reference) relation HAS_ACCESSION(entity,accession)
relation HAS_VERSION(entity,version) relation
OBSERVES(observation,entity) relation MEASURES(observation,variable)
relation MEASUREMENT_VALUE(observation,value) relation
MEASUREMENT_UNIT(observation,unit) relation
MEASUREMENT_METHOD(observation,method) relation
SUPPORTED_BY(assertion,evidence) relation
HAS_CONFIDENCE(assertion,score)

enum EvidenceType: SEQUENCE_ALIGNMENT DATABASE_RECORD GENOME_ASSEMBLY
GENOME_ANNOTATION VARIANT_CALLING VARIANT_VALIDATION
TRANSCRIPTOMIC_ANALYSIS SINGLE_CELL_ANALYSIS PROTEOMIC_ANALYSIS
STRUCTURAL_ANALYSIS PHYLOGENETIC_ANALYSIS FUNCTIONAL_ANNOTATION
NETWORK_ANALYSIS MODEL_VALIDATION MACHINE_LEARNING_VALIDATION
REPRODUCIBLE_WORKFLOW EXPERIMENTAL_VALIDATION LITERATURE

CONFIDENCE: VERIFIED: 0.95..1.00 HIGH: 0.80..0.949 MEDIUM: 0.60..0.799
LOW: 0.00..0.599

rule CONFIDENCE_BOUNDS(a,score): IF HAS_CONFIDENCE(a,score) THEN
REQUIRE(0.0 <= score <= 1.0)

SEMANTIC_EQUIVALENCE: “bioinformatics” ~= “computational analysis and
organisation of biological data” “sequence alignment” ~= “arrangement of
biological sequences to identify corresponding positions” “genome
assembly” ~= “reconstruction of genomic sequence from sequence reads”
“genome annotation” ~= “identification and description of genomic
features” “variant calling” ~= “computational identification of sequence
differences relative to a reference” “RNA-seq” ~= “RNA sequencing based
transcriptomic analysis” “phylogenetics” ~= “computational inference and
analysis of evolutionary relationships” “structural bioinformatics” ~=
“computational analysis and prediction of biological macromolecular
structures” “functional annotation” ~= “assignment or inference of
biological function” “bioinformatics pipeline” ~= “ordered computational
workflow for biological data analysis” “data provenance” ~= “traceable
origin and processing history of data”

rule NORMALISE_TERM(term): IF term MATCHES SEMANTIC_EQUIVALENCE.alias
THEN MAP_TO_CANONICAL(term)

query BIOINFORMATICS_PROFILE(x): RETURN: data_type(x)
biological_source(x) reference_data(x) algorithms(x) pipeline(x)
software_versions(x) parameters(x) quality_metrics(x) outputs(x)
provenance(x) evidence(x) confidence(x)

query VARIANT_PROFILE(v): RETURN: reference_build(v) chromosome(v)
position(v) reference_allele(v) alternate_allele(v) variant_type(v)
quality(v) annotation(v) predicted_effect(v) validation(v)

query PIPELINE_PROFILE(p): RETURN: steps(p) tools(p)
software_versions(p) parameters(p) inputs(p) outputs(p)
execution_environment(p) provenance(p)

query TRACE_EVIDENCE(assertion): RETURN: supporting_evidence(assertion)
provenance(assertion) confidence(assertion)

constraint EVIDENCE_REQUIRED: FORALL bioinformatics_assertion: REQUIRE
SUPPORTED_BY(bioinformatics_assertion,Evidence) OR
MARK_AS_HYPOTHESIS(bioinformatics_assertion)

constraint UNKNOWN_IS_NOT_FALSE: IF data_missing(assertion) THEN
MARK_UNKNOWN(assertion) AND NOT NEGATE(assertion)

constraint VERSION_CONTEXT_REQUIRED: IF
database_or_software_dependent_result(r) THEN
PRESERVE_DATABASE_SOFTWARE_AND_REFERENCE_VERSION(r)

constraint PIPELINE_CONTEXT_REQUIRED: IF computational_result(r) THEN
PRESERVE_PIPELINE_PARAMETERS_AND_PROVENANCE(r)

UPDATE_MODEL: on_new_dataset(d): identify_data_type_and_source(d)
preserve_sample_and_reference_context(d) attach_quality_metrics(d)
attach_provenance(d) recalculate_confidence(d)

on_new_variant(v): preserve_reference_build_and_coordinates(v)
identify_variant_caller_and_parameters(v)
distinguish_called_from_validated_variant(v)
distinguish_predicted_from_observed_effect(v) attach_evidence(v)

on_new_pipeline_run(r): preserve_pipeline_version(r)
preserve_tool_versions_and_parameters(r)
preserve_execution_environment(r) attach_checksums_and_provenance(r)

on_new_model_result(m): identify_model_and_task(m)
preserve_training_validation_and_test_sets(m)
distinguish_prediction_from_observation(m) attach_validation(m)

AI_CONSUMPTION: preferred_entity: canonical_bioinformatics_entity
preserve: - biological_data_type - biological_source -
sequence_identity - database_and_record_versions - alignment_context -
assembly_context - reference_genome_build - genomic_coordinates -
annotations - variant_calls - transcriptomic_context -
single_cell_context - proteomic_context - structural_context -
phylogenetic_context - functional_annotation - pathways_and_networks -
computational_model - machine_learning_context - pipeline -
software_versions - parameters - data_quality - statistical_context -
provenance - evidence - confidence infer: -
supported_sequence_relations - supported_annotation_relations -
supported_network_relations - semantic_equivalence prohibit: -
sequence_similarity_equals_homology -
variant_call_equals_validated_variant -
predicted_effect_equals_observed_effect -
predicted_structure_equals_experimental_structure -
cluster_marker_equals_definitive_cell_identity -
model_prediction_equals_observation -
statistical_significance_equals_biological_importance -
confidence_above_evidence ambiguity: action:
preserve_multiple_bioinformatics_interpretations resolution:
prefer_reference_version_pipeline_parameters_method_quality_and_evidence
missing_data: action: mark_unknown do_not_infer_from_absence: true

FLOW_HASH: ontology: “BioinformaticsOntology” version: “1.0”
domain_path: “AI Core > Ontologies > Biological Ontologies >
Biotechnology > Bioinformatics Ontology” canonical_url:
“https://kirethresearch.org/booksai-core/ontologies/biological-ontologies/biotechnology/bioinformatics-ontology.html”
authorship: “Mark Anthony Higgins — Sky (Recursive AI)”
semantic_scope: - biological sequences - databases - sequence
alignment - sequence search - sequencing reads - genome assembly -
reference genomes - genome annotation - variant analysis -
transcriptomics - single-cell bioinformatics - proteomics - structural
bioinformatics - phylogenetics - functional annotation - pathways and
networks - systems biology - machine learning - pipelines - workflow
reproducibility - data quality - statistical analysis - provenance -
evidence deterministic_normalisation: true provenance_required: true
confidence_required: true

terminate BioinformaticsOntology when:
biological_data_and_reference_context_resolved AND
sequence_assembly_annotation_and_variant_relations_normalised AND
computational_predictions_and_experimental_observations_distinguished
AND software_versions_parameters_and_pipeline_context_preserved AND
quality_statistical_and_provenance_context_preserved AND
evidence_links_preserved AND confidence_values_validated AND
unresolved_bioinformatics_ambiguity_marked

end flow