SCV Statements¶
Overview¶
The scv bundle section contains one record per ClinVar submission (SCV). Each record is a VA-Spec Statement that captures a submitter's interpretation of a variant — including the classification, evidence, condition, variant, and method.
SCV statements use #/ references to link to propositions, submitters, and conditions in their respective bundle sections rather than embedding full objects inline.
This section is produced by the SCV Statements procedure.
Record Structure¶
Each record is a VA-Spec Statement with the following top-level fields:
| Field | Type | Description |
|---|---|---|
id |
string | SCV accession with version — clinvar.submission:SCV000123456.1 |
type |
string | Always Statement |
proposition |
string | #/proposition<Meaning>/{id} reference to the classification proposition (the section is set by the proposition type; see ID References) |
classification |
object | MappableConcept — the submitter's classification. See Classification |
strength |
object | MappableConcept — the evidence strength. See Strength |
direction |
string | Whether the evidence supports, disputes, or is neutral toward the proposition |
quality |
object | Concept struct with conceptType: "Quality" and name (the submission level label). See Quality |
description |
string | Free-text interpretation summary (when provided by the submitter) |
contributions |
array | Submitter and date information with #/submitter/ references. See Contributions |
specifiedBy |
object | The classification method/guideline used. Its reportedIn is the method's citation Document, built like the statement's (including documentType). See Citations |
reportedIn |
array | Supporting citations — one Document per ClinVar citation: a PubMed, DOI or Bookshelf reference, or a URL-only citation such as a CIViC evidence link. A Document carries documentType when ClinVar gives a citation type. See Citations |
extensions |
array of Extension | ClinVar-specific metadata (0..*). See Extensions |
hasEvidenceLines |
array | Evidence lines for somatic clinical impact assertions. See Evidence Lines |
Classification¶
The classification field is a MappableConcept with the submitted classification:
{
"conceptType": "Classification",
"name": "Pathogenic",
"primaryCoding": {
"code": "pathogenic",
"system": "ACMG Guidelines, 2015"
},
"extensions": [
{
"name": "description",
"value": "for Hemochromatosis\nClassification is based on the criteria provided submission\nMay 2021 by Victorian Clinical Genetics Services"
}
]
}
The primaryCoding provides the machine-readable classification code and system from the clinvar_clinsig_types lookup table. The extensions array contains a human-readable description summarizing the condition, submission level, evaluation date, and submitter.
Strength¶
The strength field is a MappableConcept indicating the evidence strength:
{
"conceptType": "Strength",
"name": "Definitive",
"primaryCoding": {
"code": "definitive",
"system": "ACMG Guidelines, 2015"
}
}
The primaryCoding is present when the strength can be mapped to a specific code in the classification system.
Quality¶
The quality field is a Concept struct indicating the submission level:
| Field | Type | Description |
|---|---|---|
conceptType |
string | Always "Quality" |
name |
string | The submission level label (e.g., criteria provided, practice guideline, expert panel) |
Contributions¶
The contributions array records the submitter and key dates. Each contribution references the submitter via #/submitter/:
[
{
"type": "Contribution",
"contributor": "#/submitter/clinvar.submitter:500104",
"date": "2022-12-24",
"activityType": "submitted"
},
{
"type": "Contribution",
"contributor": "#/submitter/clinvar.submitter:500104",
"date": "2022-12-24",
"activityType": "created"
},
{
"type": "Contribution",
"contributor": "#/submitter/clinvar.submitter:500104",
"date": "2021-05-06",
"activityType": "evaluated"
}
]
Citations¶
reportedIn lists the submission's citations, and specifiedBy.reportedIn holds the citation for the assertion method. Each is a VA-Spec Document, one per ClinVar citation. A citation that ClinVar lists with several ids — ACMG 2015 has a PubMed id, a PMC id and a DOI — is a single Document:
{
"type": "Document",
"name": "ACMG, 2015",
"documentType": "general",
"pmid": "25741868",
"doi": "10.1038/gim.2015.30",
"urls": [
"https://pubmed.ncbi.nlm.nih.gov/25741868",
"https://europepmc.org/article/PMC/4544753",
"https://doi.org/10.1038/gim.2015.30"
]
}
nameis ClinVar's short label for the citation anddocumentTypeis ClinVar's citation type (general,practice guideline,review,Suggested Readingorfile in FileTrack), both verbatim and present only when ClinVar gives them. The same holds for the method citation inspecifiedBy.reportedInurlsholds every resolvable URL, ClinVar's own URL first; a citation ClinVar gives only as a URL — a CIViC evidence link, for example — is{"type": "Document", "urls": ["https://civicdb.org/links/evidence/4846"]}- A citation that is only free text has no
pmid,doiorurls. Its text is kept verbatim in theclinvarCitationTextextension. Such text is often not a reference, e.g. a list of ACMG criteria.
The pipeline never changes ClinVar's citation data silently. When an invalid id is repaired, the Document's extensions keep the original:
{
"type": "Document",
"documentType": "general",
"doi": "10.1155/2023/2564200",
"urls": ["https://doi.org/10.1155/2023/2564200"],
"extensions": [
{ "name": "originalClinvarCitation", "value": "doi:10.1155/2023/2564200" },
{ "name": "citationRepaired", "value": true }
]
}
An id that cannot be used, such as a DOI recorded as https, yields no doi or URL. It appears only as { "name": "invalidClinvarCitation", "value": "https" }. Method.reportedIn holds one Document. When ClinVar lists more than one citation for an assertion method, the first stays there and the rest are Documents in the Method's additionalClinvarMethodCitations extension. The full list of citation extensions and repair rules is on the Final Statements pipeline page.
Proposition¶
SCV propositions are delivered across the four proposition bundle sections (propositionPathogenicity, propositionOncogenicity, propositionSomaticClinicalImpact, propositionClinvarCustom) by meaning, referenced by #/proposition<Meaning>/{id}. The proposition ID combines the SCV accession with an uppercase proposition type code:
A resolved proposition contains:
| Field | Type | Description |
|---|---|---|
id |
string | Proposition ID (e.g., SCV001234567-PATH) |
type |
string | Proposition type (e.g., VariantPathogenicityProposition) |
predicate |
string | The relationship asserted (e.g., isCausalFor) |
subject |
string | #/variation/clinvar:{id} reference |
object |
string | The asserted object, as a reference. For most types: #/condition/clinvar.trait:{id} or #/conditionSet/clinvar.traitset:{id}. For VariantTherapeuticResponseProposition: #/therapy/clinvar.therapy:{sha256} (single) or #/therapyGroup/clinvar.therapygroup:{sha256} (combination) |
conditionQualifier |
string | #/condition/ or #/conditionSet/ reference — present on VariantTherapeuticResponseProposition (where object carries the therapy, so the condition moves to this qualifier) |
geneContextQualifier |
object | Gene context: the variant's one representative gene, with NCBI Gene and HGNC identifiers. Its geneContextSource extension says where the gene came from. When the variant has no single gene, a placeholder named with the submitter's symbol and no coding |
modeOfInheritanceQualifier |
object | Mode of inheritance with HPO coding (when submitted) |
penetranceQualifier |
object | Penetrance qualifier (for low-penetrance/risk factor classifications) |
Proposition Types¶
| Proposition Type | Code | Predicate | Description |
|---|---|---|---|
VariantPathogenicityProposition |
PATH |
isCausalFor |
Germline pathogenicity/benignity |
VariantOncogenicityProposition |
ONCO |
isOncogenicFor |
Oncogenicity |
VariantClinicalSignificanceProposition |
SCI |
hasClinicalSignificanceFor |
Somatic clinical impact |
ClinvarAssociationProposition |
ASSOC |
isAssociatedWith |
Association |
ClinvarRiskFactorProposition |
RF |
isRiskFactorFor |
Risk factor |
ClinvarDrugResponseProposition |
DR |
hasDrugResponseFor |
Drug response |
ClinvarProtectiveProposition |
PROTECT |
isProtectiveFor |
Protective |
ClinvarAffectsProposition |
AFF |
hasAffectFor |
Affects |
ClinvarConfersSensitivityProposition |
CS |
confersSensitivityFor |
Confers sensitivity |
ClinvarOtherProposition |
OTH |
isClinvarOtherAssociationFor |
Other |
ClinvarNotProvidedProposition |
NP |
hasNoProvidedClassificationFor |
Not provided |
ClinvarConflictingDataFromSubmitterProposition |
CDFS |
isConflictingDataFromSubmittersFor |
Conflicting data |
For somatic clinical impact, target propositions (evidence line propositions) use these codes: PROG (Prognostic), DIAG (Diagnostic), TR (Therapeutic Response).
See Propositions for the full profile documentation.
Evidence Lines¶
Evidence lines appear on somatic clinical impact (SCI) statements. Each evidence line carries a target proposition, direction, outcome, and condition extensions:
{
"type": "EvidenceLine",
"proposition": "#/propositionSomaticClinicalImpact/SCV004565358-TR",
"directionOfEvidenceProvided": "supports",
"evidenceOutcome": {
"conceptType": "Outcome",
"name": "tier 1"
},
"extensions": [ ... ]
}
The evidenceOutcome is a MappableConcept with conceptType: "Outcome" indicating the tier classification for the target proposition.
Extensions¶
Extensions carry ClinVar-specific metadata not part of the GA4GH VA-Spec statement model. Each extension follows the GA4GH Extension structure: { "name": "<name>", "value": <value> }. Extensions appear at three structural levels — on the top-level Statement, on the classification object, and on proposition qualifier objects.
See SCV Extensions (Pipeline) for details on how these extensions are built during pipeline processing.
Statement Extensions¶
| Extension Name | Value Type | Description |
|---|---|---|
clinvarScvId |
string |
The ClinVar SCV accession without version (e.g., SCV001571657). Always present. |
clinvarScvVersion |
string |
The version number of the SCV submission (e.g., 2). Always present. |
submittedCondition |
SubmittedCondition | Submitted condition provenance with normalization details, trait mapping, and #/condition/ references. Present when the SCV maps to a single condition. |
submittedConditionSet |
SubmittedConditionSet | Submitted condition set provenance with multiple condition concepts. Present when the SCV maps to multiple conditions. |
clinvarScvReviewStatus |
string |
The ClinVar review status (e.g., criteria provided, single submitter). Present when the SCV has a review status. |
submittedScvClassification |
string |
The original classification text submitted by the submitter, preserved when it differs from the normalized classification name. |
submittedScvLocalKey |
string |
The unique local key provided by the submitter for this submission. Present only when the submitter provided a local key. |
submissionLevel |
string |
The submission level code: PG, EP, CP, NOCP, NOCL, or FLAG. Present when the submission level can be determined. |
Note
Only one of submittedCondition or submittedConditionSet is present per SCV — never both. The submittedScvClassification extension is omitted when the submitted classification matches the normalized label exactly.
Classification Extensions¶
Extensions on the classification MappableConcept within the Statement.
| Extension Name | Value Type | Description |
|---|---|---|
description |
string |
A formatted multi-line description summarizing the classification context: condition name, submission level, evaluation date, and submitter name. Always present. |
The description template is: for <condition_name>\nClassification is based on the <submission_level_label> submission\n<evaluated_date> by <submitter_name>.
Qualifier Extensions¶
Extensions on proposition qualifier objects (geneContextQualifier, modeOfInheritanceQualifier, penetranceQualifier), preserving the original submitter-provided values.
| Extension Name | Value Type | Description |
|---|---|---|
submittedGeneSymbols |
array of string |
Gene symbols originally submitted by the submitter, verbatim, in submitted order. Present on geneContextQualifier when the submitter provided gene information. May differ from the normalized gene symbol. |
geneContextSource |
string |
Where the geneContextQualifier gene came from. Present on every geneContextQualifier. submittedSymbolMatches: the submitter's symbols include the qualifier's symbol. submittedSymbolDiffers: the submitter gave symbols, but none equals the qualifier's symbol; most are previous symbols of the same gene, so the genes are not necessarily different. inferredFromVariant: the submitter gave no symbols. submittedNotNormalized: the variant has no single representative gene, so the qualifier is a placeholder whose name is the submitter's symbol (several are joined with ,) and which has no primaryCoding. Symbols are compared after trimming spaces and ignoring case. The schema accepts only these four values (ClinvarGeneContextQualifier), on every proposition type. Added in 1.1.0; earlier placeholders are named submitted genes were not normalized. See the gene context FAQ. |
submittedModeOfInheritance |
string |
Mode of inheritance text as originally submitted. Present on all modeOfInheritanceQualifier objects. Preserved alongside the normalized HPO coding. |
submittedClassification |
string |
Original submitted classification text that triggered the penetrance qualifier derivation. Present on all penetranceQualifier objects. |
SubmittedCondition¶
The submittedCondition extension captures the provenance of how a single submitted condition was mapped to the normalized ClinVar trait. Present when the SCV maps to exactly one condition.
| Field | Type | Description |
|---|---|---|
condition |
string |
#/condition/clinvar.trait:{id} reference to the normalized condition in the bundle. |
conditionSet |
string |
#/conditionSet/clinvar.traitset:{id} reference (present when the RCV trait set has one trait but the reference uses the set). |
id |
string |
The submitted clinical assertion trait ID (e.g., SCV002769510.0). |
name |
string |
The submitted condition name. |
type |
string |
The submitted condition type (e.g., Disease, Finding). |
medgen_id |
string |
The submitted or resolved MedGen concept ID. |
normalized_match |
string |
#/condition/ reference to the matched normalized trait. |
normalized_resolution |
string |
The resolution method used to match the submitted condition (e.g., tm reftype xref omim, random trait assignment). |
xrefs |
array | Submitted cross-references with code and system. |
mapping |
object | The trait mapping entry used for resolution: type, ref, value. |
SubmittedConditionSet¶
The submittedConditionSet extension captures provenance for multi-condition submissions. Present when the SCV maps to more than one condition.
| Field | Type | Description |
|---|---|---|
conditionSet |
string |
#/conditionSet/clinvar.traitset:{id} reference. |
condition |
string |
#/condition/clinvar.trait:{id} reference (present when the RCV set resolves to a single trait). |
multiple_condition_explanation |
string |
ClinVar's explanation for the multi-condition grouping. |
concepts |
array | Array of individual condition provenance objects, each with the same fields as SubmittedCondition. |
Examples¶
Annotated JSONC examples of SCV statement records are available in the repository:
- SCV statement examples — pathogenicity, oncogenicity, somatic clinical impact, therapeutic response, and other statement types