Complete refactoring of variant and related annotation records

src/main/resources/avro/bdg.avdl

@@ -363,12 +363,55 @@ record NucleotideContigFragment {
  duplicated sequence.
  */
 enum StructuralVariantType {
+
+  /**
+   Breakend.  VCF INFO reserved key "SVTYPE" value "BND".
+   */
+  BREAKEND,
+
+  /**
+   Copy number variable region (may be both deletion and duplication).
+   VCF INFO reserved key "SVTYPE" value "CNV".
+   */
+  COPY_NUMBER_VARIABLE,
+
+  /**
+   Deletion relative to the reference. VCF INFO reserved key "SVTYPE" value "DEL".
+   */
   DELETION,
+
+  /**
+   Region of elevated copy number relative to the reference.
+   VCF INFO reserved key "SVTYPE" value "DUP".
+   */
+  DUPLICATION,
+
+  /**
+   Insertion of novel sequence relative to the reference.
+   VCF INFO reserved key "SVTYPE" value "INS".
+   */
   INSERTION,
+
+  /**
+   Inversion of reference sequence. VCF INFO reserved key "SVTYPE" value "INV".
+   */
   INVERSION,
+
+  /**
+   Insertion of a mobile element relative to the reference.
+   VCF INFO reserved key "SVTYPE" value "INS:ME".
+   */
   MOBILE_INSERTION,
+
+  /**
+   Deletion of mobile element relative to the reference.
+   VCF INFO reserved key "SVTYPE" value "DEL:ME".
+   */
   MOBILE_DELETION,
-  DUPLICATION,
+
+  /**
+   Tandem duplication. VCF INFO reserved key "SVTYPE" value "DUP:TANDEM".
+   */
   TANDEM_DUPLICATION
 }
 
@@ -378,19 +421,14 @@ enum StructuralVariantType {
 record StructuralVariant {
 
   /**
-   The type of this structural variant.
+   The type of this structural variant.  VCF INFO reserved key "SVTYPE".
    */
   union { null, StructuralVariantType } type = null;
 
-  /**
-   The URL of the FASTA/NucleotideContig assembly for this structural variant,
-   if one is available.
-   */
-  union { null, string } assembly = null;
-
   /**
    Whether this structural variant call has precise breakpoints or not. Default
    value is true. If the call is imprecise, confidence intervals should be provided.
+   Negation of VCF INFO reserved key "IMPRECISE".
    */
   union { boolean, null } precise = true;
 
@@ -411,34 +449,36 @@ record StructuralVariant {
 record Variant {
 
   /**
-   The Phred scaled error probability of a variant, given the probabilities of
-   the variant in a population.
-   */
-  union { null, int } variantErrorProbability = null;
-
-  /**
-   The reference contig that this variant exists on.
+   The reference contig that this variant exists on.  VCF column 1 "CONTIG".
    */
   union { null, string } contigName = null;
 
   /**
    The 0-based start position of this variant on the reference contig.
+   VCF column 2 "POS" converted to zero-based coordinate system, closed-open intervals.
    */
   union { null, long } start = null;
 
   /**
    The 0-based, exclusive end position of this variant on the reference contig.
+   Calculated by start + referenceAllele.length().
    */
   union { null, long } end = null;
 
   /**
-   A string describing the reference allele at this site.
+   Zero or more of unique names or identifiers for this variant. If this is a dbSNP variant it is
+   encouraged to use the rs number(s).  VCF column 3 "ID".
+   */
+  array<string> names = [];
+
+  /**
+   A string describing the reference allele at this site.  VCF column 4 "REF".
    */
   union { null, string } referenceAllele = null;
 
   /**
    A string describing the variant allele at this site. Should be left null if
-   the site is a structural variant.
+   the site is a structural variant.  VCF column 5 "ALT".
    */
   union { null, string } alternateAllele = null;
 
@@ -451,46 +491,65 @@ record Variant {
 
   /**
    A boolean describing whether this variant call is somatic; in this case, the
-   `referenceAllele` will have been observed in another sample.  VCF INFO header line
-   key SOMATIC.
+   `referenceAllele` will have been observed in another sample.  VCF INFO reserved
+   key "SOMATIC".
    */
   union { boolean, null } somatic = false;
 }
 
 /**
- An enumeration that describes the allele that corresponds to a genotype. Can take
- the following values:
-
- * REF: The genotype is the reference allele
- * ALT: The genotype is the alternate allele
- * OTHER_ALT: The genotype is an unspecified other alternate allele. This occurs
-   in our schema when we have split a multi-allelic genotype into two genotype
-   records.
- * NO_CALL: The genotype could not be called.
+ An enumeration that describes the allele that corresponds to a genotype.
  */
 enum GenotypeAllele {
+
+  /**
+   The genotype is the reference allele.
+   */
   REF,
+
+  /**
+   The genotype is the alternate allele.
+   */
   ALT,
+
+  /**
+   The genotype is an unspecified other alternate allele.  This occurs in our schema
+   when we have split a multi-allelic genotype into two genotype records.
+   */
   OTHER_ALT,
+
+  /**
+   The genotype could not be called.
+   */
   NO_CALL
 }
 
 /**
- An enumeration that describes the characteristics of a genotype at a site. Can
- take the following values:
+ An enumeration that describes the characteristics of a genotype at a site.
+ */
+enum GenotypeType {
 
- * HOM_REF: All genotypes at this site were called as the reference allele.
- * HET: Genotypes at this site were called as multiple different alleles. This
+  /**
+   All genotypes at this site were called as the reference allele.
+   */
+  HOM_REF,
+
+  /**
+   Genotypes at this site were called as multiple different alleles. This
    most commonly occurs if a diploid sample's genotype contains one reference
    and one variant allele, but can also occur if the genotype contains multiple
    alternate alleles.
- * HOM_ALT: All genotypes at this site were called as a single alternate allele.
- * NO_CALL: The genotype could not be called at this site.
- */
-enum GenotypeType {
-  HOM_REF,
+   */
   HET,
+
+  /**
+   All genotypes at this site were called as a single alternate allele.
+   */
   HOM_ALT,
+
+  /**
+   The genotype could not be called at this site.
+   */
   NO_CALL
 }