Add transcript effect endpoint

Add ability to filter by allele frequency
Provide ability to get annotations by variant id
Remove transcript effects from variant annotation message

src/main/proto/ga4gh/allele_annotation_service.proto

@@ -40,6 +40,21 @@ service AlleleAnnotationService {
   // version of `VariantAnnotationSet`.
   rpc GetVariantAnnotationSet(GetVariantAnnotationSetRequest)
       returns (VariantAnnotationSet);
+
+  // Returns a list of transcript effects.
+  //
+  // `POST /transcripteffects/search` must accept a JSON version of
+  // `SearchTranscriptEffectsRequest` as the post body and will return a
+  // JSON version of `SearchTranscriptEffectsResponse`.
+  rpc SearchTranscriptEffects(SearchTranscriptEffectsRequest)
+      returns (SearchTranscriptEffectsResponse);
+
+  // Gets a `Transcript Effect` by ID.
+  //
+  // `GET /transcripteffects/{transcript_effect_id}` will return a JSON
+  // version of `TranscriptEffect`.
+  rpc GetTranscriptEffect(GetTranscriptEffectRequest)
+      returns (TranscriptEffect);
 }
 
 /******************  /variantannotations  *********************/
@@ -48,46 +63,34 @@ message SearchVariantAnnotationsRequest {
   // Required. The ID of the variant annotation set to search over.
   string variant_annotation_set_id = 1;
 
+  // If a variant_id is provided, only return annotations for that variant.
+  // Either a variant_id or range is required.
+  string variant_id = 9;
+
   // Only return variants with reference alleles on the reference with this
-  // name. One of this field or `reference_id` is required.
+  // name. One of this field or `reference_id` is required if specifying a range.
   string reference_name = 2;
 
   // Only return variants with reference alleles on the reference with this
-  // ID. One of this field or `reference_name` is required.
+  // ID. One of this field or `reference_name` is required if specifying a range.
   string reference_id = 3;
 
-  // Required if reference_name or reference_id supplied. The beginning of the
+  // Required if specifying a range. The beginning of the
   // window (0-based, inclusive) for which variants with overlapping reference
   // alleles should be returned. Genomic positions are non-negative integers
   // less than reference length. Requests spanning the join of circular
   // genomes are represented as two requests one on each side of the join
   // (position 0).
   int64 start = 4;
 
-  // Required if reference_name or reference_id supplied. The end of the window
+  // Required if specifying a range. The end of the window
   // (0-based, exclusive) for which variants with overlapping reference
   // alleles should be returned.
   int64 end = 5;
-
-  // This section will be re-instated when features are available in the API
-  //
-  // Only return variant annotations for any of these features.
-  // Features may include specific transcripts or genes. A search by gene will
-  // return information for all transcripts associated with the gene in the
-  // variant annotation set.
-  // This or a location (referenceName/referenceId plus optional start and end)
-  // must be supplied.
-  // If empty, return all variant annotations in specified window.
-  // repeated string feature_ids;
-
-  // This filter allows variant, transcript combinations to be extracted by
-  // effect type(s). Only return variant annotations including any of these
-  // effects and only return transcript effects including any of these
-  // effects. Exact matching across all fields of the Sequence Ontology
-  // OntologyTerm is required. (A transcript effect may have multiple SO
-  // effects which will all be reported.) If empty, return all variant
-  // annotations.
-  repeated OntologyTerm effects = 6;
+
+  // Only return variants with an allele frequency annotation over this value.
+  // Return all variants by default.
+  double frequency_threshold = 10;
 
   // Specifies the maximum number of results to return in a single page. If
   // unspecified, a system default will be used.
@@ -145,3 +148,81 @@ message GetVariantAnnotationSetRequest {
   // The ID of the `VariantAnnotationSet` to be retrieved.
   string variant_annotation_set_id = 1;
 }
+
+/******************  /transcripteffects  *********************/
+// This request maps to the body of `POST /transcripteffects/search` as
+// JSON.
+message SearchTranscriptEffectsRequest {
+  // Required. The `VariantAnnotationSet` to search.
+  string variant_annotation_set_id = 1;
+
+  // If a variant_annotation_id is provided only return transcript effects
+  // from this annotation.
+  // Either variant_annotation_id feature_id, or range is required.
+  string variant_annotation_id = 2;
+
+  // Only return variants with reference alleles on the reference with this
+  // name. One of this field or `reference_id` is required if specifying a range.
+  string reference_name = 3;
+
+  // Only return variants with reference alleles on the reference with this
+  // ID. One of this field or `reference_name` is required if specifying a range.
+  string reference_id = 4;
+
+  // Required if specifying a range. The beginning of the
+  // window (0-based, inclusive) for which variants with overlapping reference
+  // alleles should be returned. Genomic positions are non-negative integers
+  // less than reference length. Requests spanning the join of circular
+  // genomes are represented as two requests one on each side of the join
+  // (position 0).
+  int64 start = 5;
+
+  // Required if specifying a range. The end of the window
+  // (0-based, exclusive) for which variants with overlapping reference
+  // alleles should be returned.
+  int64 end = 6;
+
+  // Only return transcript effects for any of these features.
+  // Features may include specific transcripts or genes. A search by gene will
+  // return information for all transcripts associated with the gene in the
+  // variant annotation set.
+  // This, a range, or feature ID are required.
+  // If empty, return all variant annotations in specified window.
+  repeated string feature_ids = 7;
+
+  // Only return transcript effects including any of these
+  // effects. Exact matching across all fields of the Sequence Ontology
+  // OntologyTerm is required. (A transcript effect may have multiple SO
+  // effects which will all be reported.) If empty, transcript effects.
+  repeated OntologyTerm effects = 8;
+
+  // Only return transcript effects with these alternate bases.
+  string alternate_bases = 9;
+
+  // Specifies the maximum number of results to return in a single page. If
+  // unspecified, a system default will be used.
+  int32 page_size = 10;
+
+  // The continuation token, which is used to page through large result sets.
+  // To get the next page of results, set this parameter to the value of
+  // `next_page_token` from the previous response.
+  string page_token = 11;
+}
+
+// This is the response from `POST /transcripteffects/search` expressed as
+// JSON.
+message SearchTranscriptEffectsResponse {
+  // The list of matching variant annotation sets.
+  repeated TranscriptEffect transcript_effects = 1;
+
+  // The continuation token, which is used to page through large result sets.
+  // Provide this value in a subsequent request to return the next page of
+  // results. This field will be empty if there aren't any additional results.
+  string next_page_token = 2;
+}
+
+// This request maps to the URL `GET /transcripteffects/{id}`.
+message GetTranscriptEffectRequest {
+  // The ID of the `TranscriptEffect` to be retrieved.
+  string transcript_effect_id = 1;
+}

src/main/proto/ga4gh/allele_annotations.proto

@@ -93,6 +93,12 @@ message TranscriptEffect {
   //  The ID of the transcript effect record
   string id = 1;
 
+  // The ID of the variant annotation this transcript effect belongs to.
+  string variant_annotation_id = 11;
+
+  // The ID of the variant this transcript effect's annotation belongs to.
+  string variant_id = 12;
+
   // TODO: derive unique id from digest of data [location, allele, transcript?]
 
   // The id of the transcript feature the annotation is relative to.
@@ -138,10 +144,9 @@ message VariantAnnotation {
 
   // The time at which this record was created, in ISO 8601 format.
   string created = 4;
-
-  // The transcript effect annotation for the alleles of this variant. Each
-  // one represents the effect of a single allele on a single transcript.
-  repeated TranscriptEffect transcript_effects = 5;
+
+  // The allele frequency provided by the annotation.
+  double allele_frequency = 7;
 
   // Additional annotation data in key-value pairs.
   map<string, google.protobuf.ListValue> info = 6;