{
  "status": "Exploratory independent-source transfer only: Pangolin aggregation provenance incomplete",
  "source": "https://www.nature.com/articles/s41467-024-52474-4",
  "sourceFiles": [
    "41467_2024_52474_MOESM6_ESM.csv",
    "41467_2024_52474_MOESM7_ESM.csv"
  ],
  "totalAssayRows": 468,
  "eligibleAssayRows": 366,
  "uniqueEligibleVariants": 224,
  "eligibleByAssay": {
    "HEK": 190,
    "iPSC-CM": 176
  },
  "rejectionCountsNonexclusive": {
    "non_SNV": 18,
    "indeterminate_or_missing_assay": 87,
    "missing_or_invalid_predictor": 6,
    "missing_location": 0,
    "missing_coordinate": 0,
    "reference_mismatch_or_nonSNV": 13,
    "historical_MFASS_variant_overlap": 0
  },
  "exonsByAssay": {
    "HEK": 19,
    "iPSC-CM": 19
  },
  "referenceMatchedEligibleRows": 366,
  "referenceURL": "https://api.genome.ucsc.edu/getData/sequence?genome=hg38;chrom=chr3;start=38554271;end=38620972",
  "referenceSHA256": "a195aeb665a34ec2d76a44c0c1997291833626dcc9b1e90aaea1aa364f233f50",
  "historicalMFASSVariantOverlaps": 0,
  "v3TrainingGeneOverlap": [],
  "freshGene": "SCN5A",
  "scoreDefinition": "S0=max(AG,AL,DG,DL) only when all components present; P0=published Pangolin column, unchanged. Its aggregation implementation not deposited; cannot verify equivalence to development max-delta scores.",
  "secondaryAssayDependence": "HEK and iPSC-CM share SCN5A variants/exons and do not constitute two independent studies.",
  "clinicalIndependence": "Only published assay Outcome is used. Clinical classifications and protein function not used as target labels.",
  "selectionBias": "Some variants chosen with high SpliceAI scores; cannot extrapolate to unselected variants.",
  "sourceSHA256": {
    "41467_2024_52474_MOESM6_ESM.csv": "607814b68c05288e31ba221c964bee22daba0909d294beda55d307155246a06a",
    "41467_2024_52474_MOESM7_ESM.csv": "5b78c17cb06a737ff895cf2344bfa4133ce68dc4857c8d657c63da3803b4fd17"
  }
}