Clinvar Submitted Tools#

Configuration File: clinvar_submitted_tools.json Tool Type: Local Tools Count: 2

This page contains all tools defined in the clinvar_submitted_tools.json configuration file.

Available Tools#

ClinVar_get_submitted_records (Type: ClinVarSubmittedRecordsTool)#

Retrieve the individual per-submitter assertions (SCV / ClinicalAssertion records) for a ClinVar …

ClinVar_get_submitted_records tool specification

Tool Information:

  • Name: ClinVar_get_submitted_records

  • Type: ClinVarSubmittedRecordsTool

  • Description: Retrieve the individual per-submitter assertions (SCV / ClinicalAssertion records) for a ClinVar variant. Unlike ClinVar_search_variants / ClinVar_get_variant_details / ClinVar_get_clinical_significance (which use esummary and return only the single AGGREGATE germline classification), this tool calls NCBI eutils efetch (rettype=vcv) and parses the full VariationArchive XML to expose every submitter’s own classification, review status, condition and last-evaluated date. Use this when you need to see who classified a variant and how (e.g. to inspect conflicting interpretations behind a ‘Conflicting classifications of pathogenicity’ aggregate). Keyless NCBI endpoint.

Parameters:

  • variant_id (string) (required) ClinVar variant identifier. Accepts a VCV accession (e.g. ‘VCV000013961’) OR a bare ClinVar variation id (e.g. ‘13961’); a numeric id is normalized to VCV%09d. Example: ‘VCV000013961’ (BRAF V600E).

Example Usage:

query = {
    "name": "ClinVar_get_submitted_records",
    "arguments": {
        "variant_id": "example_value"
    }
}
result = tu.run(query)

ClinVar_search_by_region (Type: ClinVarSearchByRegion)#

Find ClinVar variants that OVERLAP a genomic region, including large copy-number variants that be…

ClinVar_search_by_region tool specification

Tool Information:

  • Name: ClinVar_search_by_region

  • Type: ClinVarSearchByRegion

  • Description: Find ClinVar variants that OVERLAP a genomic region, including large copy-number variants that begin far outside it. Use this for ‘which variant is located at chr:start-end’ questions: Entrez’s chrpos search matches a variant’s START position, so a narrow window misses a pathogenic CNV that spans the region but starts megabases upstream (an 11 bp search at chr7:155593770-155593780 returns nothing, though a 1.5 Mb pathogenic loss covers it). The search therefore widens the window upstream by ‘margin’ and keeps only records whose span genuinely overlaps; records Entrez indexes as single-base are excluded from the upstream part of that window, since a one-base variant starting outside the region cannot reach into it. Results are sorted smallest span first, so the most specific overlapping variant comes first. Completeness is reported, not assumed: ‘total_available’ is Entrez’s own count for the search, ‘n_candidates’ is how many of those records were actually fetched and checked for overlap, and ‘truncated’ is true (with a ‘truncation_note’) when the second number is smaller than the first – in that case an empty or short ‘variants’ list is NOT evidence that nothing overlaps the region.

Parameters:

  • region (string) (optional) Region as written, e.g. ‘chr7:155593770-155593780’. Alternative to chrom/start/end.

  • chrom (string) (optional) Chromosome, e.g. ‘7’ or ‘chr7’.

  • start (integer) (optional) Region start (1-based).

  • end (integer) (optional) Region end (1-based, inclusive).

  • assembly (string) (optional) Assembly the coordinates are in. Default GRCh37.

  • clinical_significance (string) (optional) Optional filter, e.g. ‘pathogenic’.

  • margin (integer) (optional) How far upstream to look for variants that start before the region and span into it. Default 2000000. Raising it costs few extra candidates because single-base records upstream are filtered out, so it is a safe way to reach larger spanning CNVs – unless ‘truncated’ is already true, in which case a wider window only makes the truncation worse.

  • max_results (integer) (optional) Maximum overlapping variants to return; the full number found is reported as ‘n_overlapping’. Default 50.

Example Usage:

query = {
    "name": "ClinVar_search_by_region",
    "arguments": {
    }
}
result = tu.run(query)