// Copyright 2024 Google LLC
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

syntax = "proto3";

package google.genomics.v1;

import "google/api/annotations.proto";
import "google/genomics/v1/range.proto";
import "google/genomics/v1/readalignment.proto";
import "google/genomics/v1/readgroupset.proto";
import "google/longrunning/operations.proto";
import "google/protobuf/empty.proto";
import "google/protobuf/field_mask.proto";

option cc_enable_arenas = true;
option go_package = "google.golang.org/genproto/googleapis/genomics/v1;genomics";
option java_multiple_files = true;
option java_outer_classname = "ReadsProto";
option java_package = "com.google.genomics.v1";

service StreamingReadService {
  // Returns a stream of all the reads matching the search request, ordered
  // by reference name, position, and ID.
  rpc StreamReads(StreamReadsRequest) returns (stream StreamReadsResponse) {
    option (google.api.http) = {
      post: "/v1/reads:stream"
      body: "*"
    };
  }
}

// The Readstore. A data store for DNA sequencing Reads.
service ReadServiceV1 {
  // Creates read group sets by asynchronously importing the provided
  // information.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  //
  // The caller must have WRITE permissions to the dataset.
  //
  // ## Notes on [BAM](https://samtools.github.io/hts-specs/SAMv1.pdf) import
  //
  // - Tags will be converted to strings - tag types are not preserved
  // - Comments (`@CO`) in the input file header will not be preserved
  // - Original header order of references (`@SQ`) will not be preserved
  // - Any reverse stranded unmapped reads will be reverse complemented, and
  // their qualities (also the "BQ" and "OQ" tags, if any) will be reversed
  // - Unmapped reads will be stripped of positional information (reference name
  // and position)
  rpc ImportReadGroupSets(ImportReadGroupSetsRequest)
      returns (google.longrunning.Operation) {
    option (google.api.http) = {
      post: "/v1/readgroupsets:import"
      body: "*"
    };
  }

  // Exports a read group set to a BAM file in Google Cloud Storage.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  //
  // Note that currently there may be some differences between exported BAM
  // files and the original BAM file at the time of import. See
  // [ImportReadGroupSets][google.genomics.v1.ReadServiceV1.ImportReadGroupSets]
  // for caveats.
  rpc ExportReadGroupSet(ExportReadGroupSetRequest)
      returns (google.longrunning.Operation) {
    option (google.api.http) = {
      post: "/v1/readgroupsets/{read_group_set_id}:export"
      body: "*"
    };
  }

  // Searches for read group sets matching the criteria.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  //
  // Implements
  // [GlobalAllianceApi.searchReadGroupSets](https://github.com/ga4gh/schemas/blob/v0.5.1/src/main/resources/avro/readmethods.avdl#L135).
  rpc SearchReadGroupSets(SearchReadGroupSetsRequest)
      returns (SearchReadGroupSetsResponse) {
    option (google.api.http) = {
      post: "/v1/readgroupsets/search"
      body: "*"
    };
  }

  // Updates a read group set.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  //
  // This method supports patch semantics.
  rpc UpdateReadGroupSet(UpdateReadGroupSetRequest) returns (ReadGroupSet) {
    option (google.api.http) = {
      patch: "/v1/readgroupsets/{read_group_set_id}"
      body: "read_group_set"
    };
  }

  // Deletes a read group set.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  rpc DeleteReadGroupSet(DeleteReadGroupSetRequest)
      returns (google.protobuf.Empty) {
    option (google.api.http) = {
      delete: "/v1/readgroupsets/{read_group_set_id}"
    };
  }

  // Gets a read group set by ID.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  rpc GetReadGroupSet(GetReadGroupSetRequest) returns (ReadGroupSet) {
    option (google.api.http) = {
      get: "/v1/readgroupsets/{read_group_set_id}"
    };
  }

  // Lists fixed width coverage buckets for a read group set, each of which
  // correspond to a range of a reference sequence. Each bucket summarizes
  // coverage information across its corresponding genomic range.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  //
  // Coverage is defined as the number of reads which are aligned to a given
  // base in the reference sequence. Coverage buckets are available at several
  // precomputed bucket widths, enabling retrieval of various coverage 'zoom
  // levels'. The caller must have READ permissions for the target read group
  // set.
  rpc ListCoverageBuckets(ListCoverageBucketsRequest)
      returns (ListCoverageBucketsResponse) {
    option (google.api.http) = {
      get: "/v1/readgroupsets/{read_group_set_id}/coveragebuckets"
    };
  }

  // Gets a list of reads for one or more read group sets.
  //
  // For the definitions of read group sets and other genomics resources, see
  // [Fundamentals of Google
  // Genomics](https://cloud.google.com/genomics/fundamentals-of-google-genomics)
  //
  // Reads search operates over a genomic coordinate space of reference sequence
  // & position defined over the reference sequences to which the requested
  // read group sets are aligned.
  //
  // If a target positional range is specified, search returns all reads whose
  // alignment to the reference genome overlap the range. A query which
  // specifies only read group set IDs yields all reads in those read group
  // sets, including unmapped reads.
  //
  // All reads returned (including reads on subsequent pages) are ordered by
  // genomic coordinate (by reference sequence, then position). Reads with
  // equivalent genomic coordinates are returned in an unspecified order. This
  // order is consistent, such that two queries for the same content (regardless
  // of page size) yield reads in the same order across their respective streams
  // of paginated responses.
  //
  // Implements
  // [GlobalAllianceApi.searchReads](https://github.com/ga4gh/schemas/blob/v0.5.1/src/main/resources/avro/readmethods.avdl#L85).
  rpc SearchReads(SearchReadsRequest) returns (SearchReadsResponse) {
    option (google.api.http) = {
      post: "/v1/reads/search"
      body: "*"
    };
  }
}

// The read group set search request.
message SearchReadGroupSetsRequest {
  // Restricts this query to read group sets within the given datasets. At least
  // one ID must be provided.
  repeated string dataset_ids = 1;

  // Only return read group sets for which a substring of the name matches this
  // string.
  string name = 3;

  // 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
  // `nextPageToken` from the previous response.
  string page_token = 2;

  // The maximum number of results to return in a single page. If unspecified,
  // defaults to 256. The maximum value is 1024.
  int32 page_size = 4;
}

// The read group set search response.
message SearchReadGroupSetsResponse {
  // The list of matching read group sets.
  repeated ReadGroupSet read_group_sets = 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;
}

// The read group set import request.
message ImportReadGroupSetsRequest {
  enum PartitionStrategy {
    PARTITION_STRATEGY_UNSPECIFIED = 0;

    // In most cases, this strategy yields one read group set per file. This is
    // the default behavior.
    //
    // Allocate one read group set per file per sample. For BAM files, read
    // groups are considered to share a sample if they have identical sample
    // names. Furthermore, all reads for each file which do not belong to a read
    // group, if any, will be grouped into a single read group set per-file.
    PER_FILE_PER_SAMPLE = 1;

    // Includes all read groups in all imported files into a single read group
    // set. Requires that the headers for all imported files are equivalent. All
    // reads which do not belong to a read group, if any, will be grouped into a
    // separate read group set.
    MERGE_ALL = 2;
  }

  // Required. The ID of the dataset these read group sets will belong to. The
  // caller must have WRITE permissions to this dataset.
  string dataset_id = 1;

  // The reference set to which the imported read group sets are aligned to, if
  // any. The reference names of this reference set must be a superset of those
  // found in the imported file headers. If no reference set id is provided, a
  // best effort is made to associate with a matching reference set.
  string reference_set_id = 4;

  // A list of URIs pointing at [BAM
  // files](https://samtools.github.io/hts-specs/SAMv1.pdf)
  // in Google Cloud Storage.
  // Those URIs can include wildcards (*), but do not add or remove
  // matching files before import has completed.
  //
  // Note that Google Cloud Storage object listing is only eventually
  // consistent: files added may be not be immediately visible to
  // everyone. Thus, if using a wildcard it is preferable not to start
  // the import immediately after the files are created.
  repeated string source_uris = 2;

  // The partition strategy describes how read groups are partitioned into read
  // group sets.
  PartitionStrategy partition_strategy = 5;
}

// The read group set import response.
message ImportReadGroupSetsResponse {
  // IDs of the read group sets that were created.
  repeated string read_group_set_ids = 1;
}

// The read group set export request.
message ExportReadGroupSetRequest {
  // Required. The Google Cloud project ID that owns this
  // export. The caller must have WRITE access to this project.
  string project_id = 1;

  // Required. A Google Cloud Storage URI for the exported BAM file.
  // The currently authenticated user must have write access to the new file.
  // An error will be returned if the URI already contains data.
  string export_uri = 2;

  // Required. The ID of the read group set to export. The caller must have
  // READ access to this read group set.
  string read_group_set_id = 3;

  // The reference names to export. If this is not specified, all reference
  // sequences, including unmapped reads, are exported.
  // Use `*` to export only unmapped reads.
  repeated string reference_names = 4;
}

message UpdateReadGroupSetRequest {
  // The ID of the read group set to be updated. The caller must have WRITE
  // permissions to the dataset associated with this read group set.
  string read_group_set_id = 1;

  // The new read group set data. See `updateMask` for details on mutability of
  // fields.
  ReadGroupSet read_group_set = 2;

  // An optional mask specifying which fields to update. Supported fields:
  //
  // * [name][google.genomics.v1.ReadGroupSet.name].
  // * [referenceSetId][google.genomics.v1.ReadGroupSet.reference_set_id].
  //
  // Leaving `updateMask` unset is equivalent to specifying all mutable
  // fields.
  google.protobuf.FieldMask update_mask = 3;
}

message DeleteReadGroupSetRequest {
  // The ID of the read group set to be deleted. The caller must have WRITE
  // permissions to the dataset associated with this read group set.
  string read_group_set_id = 1;
}

message GetReadGroupSetRequest {
  // The ID of the read group set.
  string read_group_set_id = 1;
}

message ListCoverageBucketsRequest {
  // Required. The ID of the read group set over which coverage is requested.
  string read_group_set_id = 1;

  // The name of the reference to query, within the reference set associated
  // with this query. Optional.
  string reference_name = 3;

  // The start position of the range on the reference, 0-based inclusive. If
  // specified, `referenceName` must also be specified. Defaults to 0.
  int64 start = 4;

  // The end position of the range on the reference, 0-based exclusive. If
  // specified, `referenceName` must also be specified. If unset or 0, defaults
  // to the length of the reference.
  int64 end = 5;

  // The desired width of each reported coverage bucket in base pairs. This
  // will be rounded down to the nearest precomputed bucket width; the value
  // of which is returned as `bucketWidth` in the response. Defaults
  // to infinity (each bucket spans an entire reference sequence) or the length
  // of the target range, if specified. The smallest precomputed
  // `bucketWidth` is currently 2048 base pairs; this is subject to
  // change.
  int64 target_bucket_width = 6;

  // 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
  // `nextPageToken` from the previous response.
  string page_token = 7;

  // The maximum number of results to return in a single page. If unspecified,
  // defaults to 1024. The maximum value is 2048.
  int32 page_size = 8;
}

// A bucket over which read coverage has been precomputed. A bucket corresponds
// to a specific range of the reference sequence.
message CoverageBucket {
  // The genomic coordinate range spanned by this bucket.
  Range range = 1;

  // The average number of reads which are aligned to each individual
  // reference base in this bucket.
  float mean_coverage = 2;
}

message ListCoverageBucketsResponse {
  // The length of each coverage bucket in base pairs. Note that buckets at the
  // end of a reference sequence may be shorter. This value is omitted if the
  // bucket width is infinity (the default behaviour, with no range or
  // `targetBucketWidth`).
  int64 bucket_width = 1;

  // The coverage buckets. The list of buckets is sparse; a bucket with 0
  // overlapping reads is not returned. A bucket never crosses more than one
  // reference sequence. Each bucket has width `bucketWidth`, unless
  // its end is the end of the reference sequence.
  repeated CoverageBucket coverage_buckets = 2;

  // 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 = 3;
}

// The read search request.
message SearchReadsRequest {
  // The IDs of the read groups sets within which to search for reads. All
  // specified read group sets must be aligned against a common set of reference
  // sequences; this defines the genomic coordinates for the query. Must specify
  // one of `readGroupSetIds` or `readGroupIds`.
  repeated string read_group_set_ids = 1;

  // The IDs of the read groups within which to search for reads. All specified
  // read groups must belong to the same read group sets. Must specify one of
  // `readGroupSetIds` or `readGroupIds`.
  repeated string read_group_ids = 5;

  // The reference sequence name, for example `chr1`, `1`, or `chrX`. If set to
  // `*`, only unmapped reads are returned. If unspecified, all reads (mapped
  // and unmapped) are returned.
  string reference_name = 7;

  // The start position of the range on the reference, 0-based inclusive. If
  // specified, `referenceName` must also be specified.
  int64 start = 8;

  // The end position of the range on the reference, 0-based exclusive. If
  // specified, `referenceName` must also be specified.
  int64 end = 9;

  // 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
  // `nextPageToken` from the previous response.
  string page_token = 3;

  // The maximum number of results to return in a single page. If unspecified,
  // defaults to 256. The maximum value is 2048.
  int32 page_size = 4;
}

// The read search response.
message SearchReadsResponse {
  // The list of matching alignments sorted by mapped genomic coordinate,
  // if any, ascending in position within the same reference. Unmapped reads,
  // which have no position, are returned contiguously and are sorted in
  // ascending lexicographic order by fragment name.
  repeated Read alignments = 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;
}

// The stream reads request.
message StreamReadsRequest {
  // The Google Cloud project ID which will be billed
  // for this access. The caller must have WRITE access to this project.
  // Required.
  string project_id = 1;

  // The ID of the read group set from which to stream reads.
  string read_group_set_id = 2;

  // The reference sequence name, for example `chr1`,
  // `1`, or `chrX`. If set to *, only unmapped reads are
  // returned.
  string reference_name = 3;

  // The start position of the range on the reference, 0-based inclusive. If
  // specified, `referenceName` must also be specified.
  int64 start = 4;

  // The end position of the range on the reference, 0-based exclusive. If
  // specified, `referenceName` must also be specified.
  int64 end = 5;

  // Restricts results to a shard containing approximately `1/totalShards`
  // of the normal response payload for this query. Results from a sharded
  // request are disjoint from those returned by all queries which differ only
  // in their shard parameter. A shard may yield 0 results; this is especially
  // likely for large values of `totalShards`.
  //
  // Valid values are `[0, totalShards)`.
  int32 shard = 6;

  // Specifying `totalShards` causes a disjoint subset of the normal response
  // payload to be returned for each query with a unique `shard` parameter
  // specified. A best effort is made to yield equally sized shards. Sharding
  // can be used to distribute processing amongst workers, where each worker is
  // assigned a unique `shard` number and all workers specify the same
  // `totalShards` number. The union of reads returned for all sharded queries
  // `[0, totalShards)` is equal to those returned by a single unsharded query.
  //
  // Queries for different values of `totalShards` with common divisors will
  // share shard boundaries. For example, streaming `shard` 2 of 5
  // `totalShards` yields the same results as streaming `shard`s 4 and 5 of 10
  // `totalShards`. This property can be leveraged for adaptive retries.
  int32 total_shards = 7;
}

message StreamReadsResponse {
  repeated Read alignments = 1;
}