feat!: refactory query schema to use protobuf (#374)

Cargo.toml

@@ -12,7 +12,7 @@ rust-version = "1.70.0"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-annonars = "0.39"
+annonars = "0.40"
 anyhow = "1.0"
 async-compression = { version = "0.4", features = ["tokio", "gzip"] }
 aws-sdk-s3 = { version = "1.38", features = ["behavior-version-latest"] }
@@ -35,16 +35,16 @@ ext-sort = { version = "0.1", features = ["memory-limit", "bytesize"] }
 fastrand = "2.0"
 flate2 = "1.0"
 futures = "0.3.30"
-hgvs = "0.16"
+hgvs = "0.17"
 indexmap = { version = "2.2", features = ["serde"] }
 itertools = "0.13"
 log = "0.4"
-mehari = "0.25.6"
+mehari = "0.26.0"
 multimap = "0.10"
 pbjson = "0.6"
 pbjson-types = "0.6"
 procfs = "0.16"
-prost = "0.12"
+prost = "0.13.1"
 rand = "0.8"
 rand_core = "0.6"
 rayon = "1.10"
@@ -66,7 +66,7 @@ tracing-subscriber = "0.3"
 uuid = { version = "1.9", features = ["v4", "fast-rng", "serde"] }
 
 [dependencies.noodles]
-version = "0.76.0"
+version = "0.77.0"
 features = ["bgzf", "core", "csi", "tabix", "vcf", "bcf"]
 
 

build.rs

@@ -9,6 +9,7 @@ fn main() -> Result<(), anyhow::Error> {
         "varfish/v1/clinvar.proto",
         "varfish/v1/sv.proto",
         "varfish/v1/worker.proto",
+        "varfish/v1/seqvars/query.proto",
     ]
     .iter()
     .map(|f| root.join(f))

protos/varfish/v1/seqvars/query.proto

@@ -0,0 +1,343 @@
+syntax = "proto3";
+
+package varfish.v1.seqvars.query;
+
+// Enumeration for recessive mode queries.
+enum RecessiveMode {
+  // Unknown recessive mode.
+  RECESSIVE_MODE_UNSPECIFIED = 0;
+  // Disabled recessive mode.
+  RECESSIVE_MODE_DISABLED = 1;
+  // Compound heterozygous recessive mode.
+  RECESSIVE_MODE_COMPOUND_HETEROZYGOUS = 2;
+  // Homozygous recessive mode.
+  RECESSIVE_MODE_HOMOZYGOUS = 3;
+  // Generic recessive mode.
+  RECESSIVE_MODE_ANY = 4;
+}
+
+// Choice for genotype.
+enum GenotypeChoice {
+  // Unknown genotype.
+  GENOTYPE_CHOICE_UNSPECIFIED = 0;
+  // Any genoype.
+  GENOTYPE_CHOICE_ANY = 1;
+  // Reference genotype.
+  GENOTYPE_CHOICE_REF = 2;
+  // Heterozygous genotype.
+  GENOTYPE_CHOICE_HET = 3;
+  // Homozygous genotype.
+  GENOTYPE_CHOICE_HOM = 4;
+  // Non-heterozygous genotype.
+  GENOTYPE_CHOICE_NON_HET = 5;
+  // Non-homozygous genotype.
+  GENOTYPE_CHOICE_NON_HOM = 6;
+  // Variant genotype.
+  GENOTYPE_CHOICE_VARIANT = 7;
+  // Recessive index.
+  GENOTYPE_CHOICE_RECESSIVE_INDEX = 8;
+  // Recessive father.
+  GENOTYPE_CHOICE_RECESSIVE_FATHER = 9;
+  // Recessive mother.
+  GENOTYPE_CHOICE_RECESSIVE_MOTHER = 10;
+}
+
+// Genotype choice for one sample.
+message SampleGenotypeChoice {
+  // Name of the sample filtered for
+  string sample = 1;
+  // Genotype choice
+  GenotypeChoice genotype = 2;
+  // Whether to include no-call (will disable quality filter).
+  bool include_no_call = 3;
+  // Whether to enable sample in filtration
+  bool enabled = 4;
+}
+
+// Genotype-related filter settings.
+message QuerySettingsGenotype {
+  // Recessive mode
+  RecessiveMode recessive_mode = 1;
+  // List of sample genotype choices
+  repeated SampleGenotypeChoice sample_genotypes = 2;
+}
+
+// Quality settings for one sample.
+message SampleQualitySettings {
+  // Name of the sample filtered for
+  string sample = 1;
+  // Drop whole variant on failure
+  bool filter_active = 2;
+  // Minimal coverage for het. sites
+  optional int32 min_dp_het = 3;
+  // Minimal coverage for hom. sites
+  optional int32 min_dp_hom = 4;
+  // Minimal genotype quality
+  optional int32 min_gq = 5;
+  // Minimal allele balance for het. variants
+  optional float min_ab = 6;
+  // Minimal number of alternate reads
+  optional int32 min_ad = 7;
+  // Maximal number of alternate reads
+  optional int32 max_ad = 8;
+}
+
+// Per-sample quality filter settings.
+message QuerySettingsQuality {
+  // List of sample quality settings
+  repeated SampleQualitySettings sample_qualities = 1;
+}
+
+// gnomAD nuclear filter options.
+message GnomadNuclearFreqyencySettings {
+  // Whether to enable filtration by 1000 Genomes.
+  bool enabled = 1;
+  // Maximal number of in-house heterozygous carriers
+  optional int32 heterozygous = 2;
+  // Maximal number of in-house homozygous carriers
+  optional int32 homozygous = 3;
+  // Maximal number of in-house hemizygous carriers
+  optional int32 hemizygous = 4;
+  // Maximal allele frequency.
+  optional float frequency = 5;
+}
+
+// gnomAD mitochondrial filter options.
+message GnomadMitochondrialFrequencySettings {
+  // Whether to enable filtration by 1000 Genomes.
+  bool enabled = 1;
+  // Maximal number of heteroplasmic carriers.
+  optional int32 heteroplasmic = 2;
+  // Maximal number of homoplasmic carriers.
+  optional int32 homoplasmic = 3;
+  // Maximal allele frequency.
+  optional float frequency = 4;
+}
+
+// HelixMtDb filter options.
+message HelixMtDbFrequencySettings {
+  // Whether to enable filtration by mtDB
+  bool enabled = 1;
+  // Maximal number of heterozygous carriers in HelixMtDb
+  optional int32 heteroplasmic = 2;
+  // Maximal number of homozygous carriers in HelixMtDb
+  optional int32 homoplasmic = 3;
+  // Maximal frequency in HelixMtDb
+  optional float frequency = 4;
+}
+
+// In-house frequency filter options.
+message InhouseFrequencySettings {
+  // Whether to enable filtration by 1000 Genomes.
+  bool enabled = 1;
+  // Maximal number of in-house heterozygous carriers
+  optional int32 heterozygous = 2;
+  // Maximal number of in-house homozygous carriers
+  optional int32 homozygous = 3;
+  // Maximal number of in-house hemizygous carriers
+  optional int32 hemizygous = 4;
+  // Maximal number of in-house carriers
+  optional int32 carriers = 5;
+}
+
+// Population frequency filter options.
+message QuerySettingsFrequency {
+  // gnomAD-exomes filter
+  GnomadNuclearFreqyencySettings gnomad_exomes = 1;
+  // gnomAD-genomes filter
+  GnomadNuclearFreqyencySettings gnomad_genomes = 2;
+  // gnomAD-MT filter
+  GnomadMitochondrialFrequencySettings gnomad_mtdna = 3;
+  // HelixMtDb filter
+  HelixMtDbFrequencySettings helixmtdb = 4;
+  // In-house filter
+  InhouseFrequencySettings inhouse = 5;
+}
+
+// The variant types.
+enum VariantType {
+  // Unknown variant type
+  VARIANT_TYPE_UNSPECIFIED = 0;
+  // SNV
+  VARIANT_TYPE_SNV = 1;
+  // Indel
+  VARIANT_TYPE_INDEL = 2;
+  // MNV
+  VARIANT_TYPE_MNV = 3;
+  // Complex Substitution
+  VARIANT_TYPE_COMPLEX_SUBSTITUTION = 4;
+}
+
+// Transcript types to consider.
+enum TranscriptType {
+  // Unknown transcript type.
+  TRANSCRIPT_TYPE_UNSPECIFIED = 0;
+  // Coding transcript.
+  TRANSCRIPT_TYPE_CODING = 1;
+  // Non-coding transcript.
+  TRANSCRIPT_TYPE_NON_CODING = 2;
+}
+
+// The Variant consequence
+enum Consequence {
+  // Unknown consequence.
+  CONSEQUENCE_UNSPECIFIED = 0;
+  /*
+   * high impact
+   */
+  // Transcript ablation.
+  CONSEQUENCE_TRANSCRIPT_ABLATION = 1;
+  // Exon loss variant.
+  CONSEQUENCE_EXON_LOSS_VARIANT = 2;
+  // Splice acceptor variant
+  CONSEQUENCE_SPLICE_ACCEPTOR_VARIANT = 3;
+  // Splice donor variant
+  CONSEQUENCE_SPLICE_DONOR_VARIANT = 4;
+  // Stop gained
+  CONSEQUENCE_STOP_GAINED = 5;
+  // Frameshift variant
+  CONSEQUENCE_FRAMESHIFT_VARIANT = 6;
+  // Stop lost
+  CONSEQUENCE_STOP_LOST = 7;
+  // Start lost
+  CONSEQUENCE_START_LOST = 8;
+  // Transcript amplification
+  CONSEQUENCE_TRANSCRIPT_AMPLIFICATION = 9;
+  /*
+   * moderate impact
+   */
+  // Disruptive inframe insertion
+  CONSEQUENCE_DISRUPTIVE_INFRAME_INSERTION = 10;
+  // Disruptive inframe deletion
+  CONSEQUENCE_DISRUPTIVE_INFRAME_DELETION = 11;
+  // Conservative inframe insertion
+  CONSEQUENCE_CONSERVATIVE_INFRAME_INSERTION = 12;
+  // Conservative inframe deletion
+  CONSEQUENCE_CONSERVATIVE_INFRAME_DELETION = 13;
+  // In-frame indel.
+  CONSEQUENCE_IN_FRAME_INDEL = 14;
+  // Missense variant
+  CONSEQUENCE_MISSENSE_VARIANT = 15;
+  /*
+   * low impact
+   */
+  // Splice donor 5th base variant.
+  CONSEQUENCE_SPLICE_DONOR_FIFTH_BASE_VARIANT = 16;
+  // Splice region variant.
+  CONSEQUENCE_SPLICE_REGION_VARIANT = 17;
+  // Splice donor region variant.
+  CONSEQUENCE_SPLICE_DONOR_REGION_VARIANT = 18;
+  // Splice polypyrimidine tract variant.
+  CONSEQUENCE_SPLICE_POLYPYRIMIDINE_TRACT_VARIANT = 19;
+  // Start retained variant.
+  CONSEQUENCE_START_RETAINED_VARIANT = 20;
+  // Stop retained variant.
+  CONSEQUENCE_STOP_RETAINED_VARIANT = 21;
+  // Synonymous variant.
+  CONSEQUENCE_SYNONYMOUS_VARIANT = 22;
+  /*
+   * modifier
+   */
+  // Coding sequence variant.
+  CONSEQUENCE_CODING_SEQUENCE_VARIANT = 23;
+  // 5' UTR exon variant.
+  CONSEQUENCE_FIVE_PRIME_UTR_EXON_VARIANT = 24;
+  // 5' UTR intron variant.
+  CONSEQUENCE_FIVE_PRIME_UTR_INTRON_VARIANT = 25;
+  // 3' UTR exon variant.
+  CONSEQUENCE_THREE_PRIME_UTR_EXON_VARIANT = 26;
+  // 3' UTR intron variant.
+  CONSEQUENCE_THREE_PRIME_UTR_INTRON_VARIANT = 27;
+  // Non-coding transcript exon variant.
+  CONSEQUENCE_NON_CODING_TRANSCRIPT_EXON_VARIANT = 28;
+  // Non-coding transcript intron variant.
+  CONSEQUENCE_NON_CODING_TRANSCRIPT_INTRON_VARIANT = 29;
+  // Upstream gene variant.
+  CONSEQUENCE_UPSTREAM_GENE_VARIANT = 30;
+  // Downstream gene variant.
+  CONSEQUENCE_DOWNSTREAM_GENE_VARIANT = 31;
+  // Intergenic variant.
+  CONSEQUENCE_INTERGENIC_VARIANT = 32;
+  // Intron variant.
+  CONSEQUENCE_INTRON_VARIANT = 33;
+}
+
+// Consequence-related filter settings.
+message QuerySettingsConsequence {
+  // The variant types.
+  repeated VariantType variant_types = 1;
+  // The transcript types.
+  repeated TranscriptType transcript_types = 2;
+  // List of consequences to consider
+  repeated Consequence consequences = 3;
+  // Maximal distance to next exon, if any
+  optional int32 max_dist_to_exon = 4;
+}
+
+// An 1-based integer range.
+message Range {
+  // 1-based start position.
+  int32 start = 1;
+  // 1-based end position.
+  int32 end = 2;
+}
+
+// Genomic region.
+message GenomicRegion {
+  // Chromosome
+  string chrom = 1;
+  // Range of region
+  optional Range range = 2;
+}
+
+// Locus-related filter settings.
+message QuerySettingsLocus {
+  // List of HGNC identifiers for filtration to genes.
+  //
+  // The server will expand gene panels to gene lists here.
+  repeated string genes = 1;
+  // List of genomic regions to limit restrict the resulting variants to
+  repeated GenomicRegion genome_regions = 2;
+}
+
+// Enumeration of canonical ClinVar germline aggregte descriptions.
+enum ClinvarGermlineAggregateDescription {
+  // Unknown description.
+  CLINVAR_GERMLINE_AGGREGATE_DESCRIPTION_UNSPECIFIED = 0;
+  // Pathogenic.
+  CLINVAR_GERMLINE_AGGREGATE_DESCRIPTION_PATHOGENIC = 1;
+  // Likely pathogenic.
+  CLINVAR_GERMLINE_AGGREGATE_DESCRIPTION_LIKELY_PATHOGENIC = 2;
+  // Uncertain significance.
+  CLINVAR_GERMLINE_AGGREGATE_DESCRIPTION_UNCERTAIN_SIGNIFICANCE = 3;
+  // Likely benign.
+  CLINVAR_GERMLINE_AGGREGATE_DESCRIPTION_LIKELY_BENIGN = 4;
+  // Benign.
+  CLINVAR_GERMLINE_AGGREGATE_DESCRIPTION_BENIGN = 5;
+}
+
+// Clinvar-related query settings.
+message QuerySettingsClinVar {
+  // Wether to require ClinVar membership
+  bool presence_required = 1;
+  // The ClinVar germline aggregate description to include.
+  repeated ClinvarGermlineAggregateDescription germline_descriptions = 2;
+  // Whether to include conflicting interpretation ClinVar variants
+  bool allow_conflicting_interpretations = 3;
+}
+
+// Store query information for one case.
+message CaseQuery {
+  // Genotype query settings.
+  QuerySettingsGenotype genotype = 1;
+  // Quality query settings.
+  QuerySettingsQuality quality = 2;
+  // Frequency query settings.
+  QuerySettingsFrequency frequency = 3;
+  // Consequence query settings.
+  QuerySettingsConsequence consequence = 4;
+  // Locus query settings.
+  QuerySettingsLocus locus = 5;
+  // ClinVar query settings.
+  QuerySettingsClinVar clinvar = 6;
+}