- Created by Nacho Medina, last modified on Apr 16, 2020
Warning
This is a work in progress documentation for v2.0.0
Overview
Genomic variant data model plays a crucial role not only in OpenCGA but also in OpenCB suite. Variant data model provides a generic way of representing any variant with any other interesting information associated with it. Variant data model is heavily used in OpenCGA when loading VCF files or when exporting query results. Variant data model is implemented in OpenCB Biodata project, this allows the rest of OpenCB projects such as CellBase to use it.
Goals
Main goals of variant data model include:
- To represent any type of variant (SNV, INDEL) or structural variant (INSERTION, DELETION, CNV, TRANSLOCATION, ...)
- To support phased variants and non-diploid organisms.
- To provide a file-format agnostic model of storing genomic variant data from VCF, gVCF, microarrays, ...
- To support rich variant annotations for researchers and clinicians
- To store sample data and stats from different datasets
- To allow customisation of variant scores and annotation
Main Features
Some of the main features of the variant data model include:
- support for any type of genomic variant
- rich and customisable annotation integrated
Data Model Overview
A high level representation of the variant looks like this, only main categories are shown:
{ "id": "1:69511:A:G", "names": ["rs75062661"], "chromosome": "1", "start": 69511, "end": 69511, "strand": "+", "length": 1, "type": "SNV", "reference": "A", "alternate": "G", "studies": [ { "studyId": "demo@family:corpasome", "secondaryAlternates": [], "files": [ { "fileId": "quartet.variants.annotated.vcf.gz" "call" : {}, "data": { "FILTER": "PASS", "QUAL": "4293.01", "HaplotypeScore": "2.4399", "MQ": "15.47", ... } }, ... ], "sampleDataKeys": ["GT", "AD", "DP", "GQ", "PL"], "samples": [ { "sampleId": "ISBM200170115", "fileIndex": 0, "data": ["1/1", "2,171", "173", "99", "2218,228,0"] }, ... ], "stats": [ { "cohortId": "ALL", "sampleCount": 4, "fileCount": 1, "alleleCount": 8, "refAlleleCount": 0, "refAlleleFreq": 0.0, "altAlleleCount": 8, "altAlleleFreq": 1.0, "genotypeCount": { "0/0": 0, "0/1": 0, "1/1": 4 }, "genotypeFreq": { "0/0": 0.0, "0/1": 0.0, "1/1": 1.0 }, "missingAlleleCount": 0, "missingGenotypeCount": 0, "maf": 0.0, "mafAllele": "A", "mgf": 0.0, "mgfGenotype": "0/0", "filterCount": { "PASS": 0, "VQSRTrancheSNP99.90to100.00": 1 }, "filterFreq": { "PASS": 0.0, "VQSRTrancheSNP99.90to100.00": 1.0 }, "qualityCount": 1, "qualityAvg": 4293.01 }, ... ], "scores": [], "issues": [] } ], "annotation": { "id": "rs2691305", "chromosome": "1", "start": 69511, "reference": "A", "alternate": "G", "hgvs": ["ENST00000335137(ENSG00000186092):c.421A>G"], "displayConsequenceType": "missense_variant", "consequenceTypes": [ { "geneName": "OR4F5", "ensemblGeneId": "ENSG00000186092", "ensemblTranscriptId": "ENST00000335137", "biotype": "protein_coding", "cdnaPosition": 421, "cdsPosition": 421, "codon": "Aca/Gca", "strand": "+", "transcriptAnnotationFlags": ["CCDS", "basic"], "exonOverlap": [ { "number": "1/1", "percentage": 0.108932465 } ], "proteinVariantAnnotation": { "uniprotAccession": "Q8NH21", "position": 141, "reference": "THR", "alternate": "ALA", "features": [ { "id": "IPR017452", "start": 34, "end": 280, "description": "GPCR, rhodopsin-like, 7TM" }, ... ], "keywords": ["Cell membrane", "Complete proteome", "Disulfide bond", ...], "substitutionScores": [ { "description": "tolerated", "score": 0.63, "source": "sift" }, { "description": "benign", "score": 0.003, "source": "polyphen" } ], }, "sequenceOntologyTerms": [ { "accession": "SO:0001583", "name": "missense_variant" } ] } ], "conservation": [ { "score": 1.149999976158142, "source": "gerp" }, { "score": 0.1289999932050705, "source": "phastCons" }, { "score": -0.527999997138977, "source": "phylop" } ], "cytoband": [ { "name": "p36.33", "chromosome": "1", "start": 1, "end": 2300000, "stain": "gneg" } ], "functionalScore": [ { "score": -0.7899999618530273, "source": "cadd_raw" }, { "score": 0.03999999910593033, "source": "cadd_scaled" } ], "geneDrugInteraction": [], "geneTraitAssociation": [], "populationFrequencies": [ { "altAllele": "G", "altAlleleFreq": 0.84222084, "altHomGenotypeFreq": 0.77478045, "hetGenotypeFreq": 0.1348808, "population": "ALL", "refAllele": "A", "refAlleleFreq": 0.15777917, "refHomGenotypeFreq": 0.090338774, "study": "GNOMAD_GENOMES" }, { "altAllele": "G", "altAlleleFreq": 0.9637507, "altHomGenotypeFreq": 0.94847214, "hetGenotypeFreq": 0.03055722, "population": "NFE", "refAllele": "A", "refAlleleFreq": 0.03624925, "refHomGenotypeFreq": 0.02097064, "study": "GNOMAD_GENOMES" } ], "repeat": [ { "chromosome": "1", "copyNumber": 2.0, "end": 87112, "id": "9119", "percentageMatch": 0.992904, "source": "genomicSuperDup", "start": 10001 }, { "chromosome": "1", "copyNumber": 2.0, "end": 87112, "id": "14903", "percentageMatch": 0.995437, "source": "genomicSuperDup", "start": 18393 } ], "traitAssociation": [ { "additionalProperties": [ { "name": "mutationSomaticStatus_in_source_file", "value": "Confirmed " "somatic " "variant"}], "alleleOrigin": [], "bibliography": [], "ethnicity": "Z", "genomicFeatures": [{"featureType": "gene", "xrefs": {"symbol": "OR4F5"}}, {"featureType": "gene", "xrefs": {"symbol": "8301"}}], "heritableTraits": [], "id": "COSM4144171", "somaticInformation": {"histologySubtype": "neoplasm", "primaryHistology": "other", "primarySite": "thyroid", "sampleSource": "", "tumourOrigin": ""}, "source": {"name": "cosmic"}, "submissions": []}]} "additionalAttributes": { "opencga": { "attribute": { "annotationId": "CURRENT", "release": "1" } } } }
Implementation
Variant data model is implemented in OpenCB Biodata project, this allows the rest of OpenCB projects such as CellBase, Oskar to
Table of Contents:
- No labels