Merge pull request #78 from Ensembl/add-genome-groups

bilalebi · web-flow · commit ddeea5dc0055 · 2025-11-20T15:49:14.000Z
Add genome groups
diff --git a/.github/workflows/deploy-docs-openapi.yaml b/.github/workflows/deploy-docs-openapi.yaml
@@ -0,0 +1,44 @@
+name: Deploy API Docs
+on:
+  push:
+    branches: [main]
+    paths: ['APISpecification.yaml']  # Only run when spec changes
+
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Create documentation page
+        run: |
+          cat > index.html << 'EOF'
+          <!DOCTYPE html>
+          <html>
+            <head>
+              <title>API Documentation</title>
+              <link rel="stylesheet" type="text/css" href="https://unpkg.com/swagger-ui-dist@3/swagger-ui.css" />
+            </head>
+            <body>
+              <div id="swagger-ui"></div>
+              <script src="https://unpkg.com/swagger-ui-dist@3/swagger-ui-bundle.js"></script>
+              <script>
+                SwaggerUIBundle({
+                  url: './APISpecification.yaml',
+                  dom_id: '#swagger-ui',
+                  presets: [
+                    SwaggerUIBundle.presets.apis,
+                    SwaggerUIBundle.presets.standalone
+                  ]
+                });
+              </script>
+            </body>
+          </html>
+          EOF
+
+      - name: Deploy to GitHub Pages
+        uses: peaceiris/actions-gh-pages@v3
+        with:
+          github_token: ${{ secrets.GITHUB_TOKEN }}
+          publish_dir: ./
+          publish_branch: gh-pages
diff --git a/.github/workflows/deploy-docs.yml b/.github/workflows/deploy-docs.yml
@@ -0,0 +1,42 @@
+name: Deploy OpenAPI docs to GitHub Pages
+
+on:
+  push:
+    branches:
+      - main
+      - add-genome-groups
+    paths:
+      - 'docs/**'
+      - '.github/workflows/deploy-docs.yml'
+
+permissions:
+  contents: read
+  pages: write
+  id-token: write
+
+concurrency:
+  group: 'pages'
+  cancel-in-progress: true
+
+jobs:
+  deploy:
+    environment:
+      name: github-pages
+      url: ${{ steps.deployment.outputs.page_url }}
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Pages
+        uses: actions/configure-pages@v5
+
+      - name: Upload artifact
+        uses: actions/upload-pages-artifact@v3
+        with:
+          path: docs
+
+      - name: Deploy to GitHub Pages
+        id: deployment
+        uses: actions/deploy-pages@v4
diff --git a/README.md b/README.md
@@ -13,4 +13,12 @@ docker-compose -f docker-compose.yml up
 ```bash
 docker-compose -f docker-compose.yml run metadata_api python -m unittest
 ```
+
+### Run locally without Docker:
+```bash
+git clone https://github.com/Ensembl/ensembl-web-metadata-api
+cd ensembl-web-metadata-api
+pip install -r requirements.txt
+PYTHONPATH='app' uvicorn main:app --host 0.0.0.0 --port 8014 --reload
+```
  
diff --git a/app/api/models/genome.py b/app/api/models/genome.py
@@ -195,4 +195,16 @@ class DatasetAttributes(BaseModel):
 class GenomeByKeyword(BaseModel):
     genome_uuid: str = Field(alias="genomeUuid", default="")
     release_version: float = Field(alias=AliasPath("release", "releaseVersion"), default=0)
-    genome_tag: str = Field(alias=AliasPath("assembly", "urlName"), default="")
+    genome_tag: str = Field(alias=AliasPath("assembly", "urlName"), default="")
+
+class GenomeGroup(BaseModel):
+    id: str = Field(alias="groupId")
+    type: str = Field(alias="groupType")
+    name: str | None = Field(alias="groupName", default=None)
+    reference_genome: BaseGenomeDetails = Field(alias="referenceGenome")
+
+class GenomeGroupsResponse(BaseModel):
+    genome_groups: list[GenomeGroup] = Field(alias="genomeGroups")
+
+class GenomesInGroupResponse(BaseModel):
+    genomes: list[BaseGenomeDetails] = Field(alias="genomes")
diff --git a/app/api/resources/grpc_client.py b/app/api/resources/grpc_client.py
@@ -169,3 +169,17 @@ def get_release(self, release_label: list[str], current_only: bool):
         request_class = self.reflector.message_class("ensembl_metadata.ReleaseRequest")
         request = request_class(release_label=release_label, current_only=current_only)
         return self.stub.GetRelease(request)
+
+    def get_genome_groups_with_reference(self, group_type: str, release_label: str):
+        request_class = self.reflector.message_class(
+            "ensembl_metadata.GroupTypeRequest"
+        )
+        request = request_class(group_type=group_type, release_label=release_label)
+        return self.stub.GetGenomeGroupsWithReference(request)
+
+    def get_genomes_in_group(self, group_id: str, release_label: str):
+        request_class = self.reflector.message_class(
+            "ensembl_metadata.GenomesInGroupRequest"
+        )
+        request = request_class(group_id=group_id, release_label=release_label)
+        return self.stub.GetGenomesInGroup(request)
diff --git a/app/api/resources/metadata.py b/app/api/resources/metadata.py
@@ -17,15 +17,16 @@
 import logging
 from typing import Annotated
 
-from fastapi import APIRouter, Request, responses, Query, HTTPException
+from fastapi import APIRouter, Request, responses, Query, Path
 from pydantic import ValidationError
 
 from api.error_response import response_error_handler
 from api.models.checksums import Checksum
 from api.models.statistics import GenomeStatistics, ExampleObjectList
 from api.models.popular_species import PopularSpeciesGroup
 from api.models.karyotype import Karyotype
-from api.models.genome import BriefGenomeDetails, GenomeDetails, DatasetAttributes, GenomeByKeyword, Release
+from api.models.genome import BriefGenomeDetails, GenomeDetails, DatasetAttributes, GenomeByKeyword, Release, \
+    GenomeGroupsResponse, GenomesInGroupResponse
 from api.models.ftplinks import FTPLinks
 from api.models.vep import VepFilePaths
 
@@ -347,3 +348,57 @@ async def get_releases(
         response_data = responses.JSONResponse(error_response, status_code=500)
 
     return response_data
+
+@router.get("/genome_groups", name="genome_groups")
+@redis_cache("genome_groups", arg_keys=["group_type", "release_label"])
+async def get_genome_groups(
+        group_type: str = Query(..., description="Group type, e.g. 'structural_variant'"),
+        release_label: str | None = Query(None, description="Optional release label, e.g. '2025-02'")
+):
+    try:
+        genome_groups_dict = MessageToDict(
+            grpc_client.get_genome_groups_with_reference(
+                group_type=group_type,
+                release_label=release_label
+            )
+        )
+        logging.debug(f"genome_groups_dict: {genome_groups_dict}")
+        if len(genome_groups_dict.get("genomeGroups", [])) == 0:
+            return response_error_handler({
+                "status": 404,
+                "details": "No genome groups found matching criteria"
+            })
+
+        genome_groups = GenomeGroupsResponse(**genome_groups_dict)
+        response_dict = genome_groups.model_dump()
+        return responses.JSONResponse(response_dict, status_code=200)
+    except Exception as ex:
+        logging.exception("Error in get_genome_groups")
+        return response_error_handler({"status": 500})
+
+@router.get("/genome_groups/{group_id}/genomes", name="genomes_in_group")
+@redis_cache("genomes_in_group", arg_keys=["group_id", "release_label"])
+async def get_genomes_in_group(
+        group_id: str = Path(..., description="Group ID, e.g. 'grch38-group'"),
+        release_label: str | None = Query(None, description="Optional release label, e.g. '2025-02'")
+):
+    try:
+        genomes_in_group_dict = MessageToDict(
+            grpc_client.get_genomes_in_group(
+                group_id=group_id,
+                release_label=release_label
+            )
+        )
+        logging.debug(f"genomes_in_group_dict: {genomes_in_group_dict}")
+        if len(genomes_in_group_dict.get("genomes", [])) == 0:
+            return response_error_handler({
+                "status": 404,
+                "details": "No genomes found in specified group"
+            })
+
+        genomes_in_group = GenomesInGroupResponse(**genomes_in_group_dict)
+        response_dict = genomes_in_group.model_dump()
+        return responses.JSONResponse(response_dict, status_code=200)
+    except Exception as ex:
+        logging.exception("Error in get_genomes_in_group")
+        return response_error_handler({"status": 500})
diff --git a/docs/APISpecification.yaml b/docs/APISpecification.yaml
@@ -382,6 +382,62 @@ paths:
         '500':
           $ref: '#/components/responses/500InternalServerError'
 
+  /api/metadata/genome_groups:
+      get:
+        summary: List genome groups with their reference genome
+        parameters:
+          - in: query
+            name: group_type
+            required: true
+            schema:
+              type: string
+          - in: query
+            name: release_label
+            required: false
+            schema:
+              type: string
+        responses:
+          "200":
+            description: OK
+            content:
+              application/json:
+                schema:
+                  type: object
+                  properties:
+                    genome_groups:
+                      type: array
+                      items:
+                        $ref: '#/components/schemas/GenomeGroup'
+          "400": { description: Bad Request }
+
+  /api/metadata/genome_groups/{group_id}/genomes:
+    get:
+      summary: List genomes in a group
+      parameters:
+        - in: path
+          name: group_id
+          required: true
+          schema:
+            type: string
+        - in: query
+          name: release_label
+          required: false
+          schema:
+            type: string
+      responses:
+        "200":
+          description: OK
+          content:
+            application/json:
+              schema:
+                type: object
+                properties:
+                  genomes:
+                    type: array
+                    items:
+                      $ref: '#/components/schemas/Genome'
+        "404": { description: Not Found }
+
 components:
   schemas:
     GenomeIDFromAssemblyAccessionResponse:
@@ -1072,6 +1128,35 @@ components:
         - name
         - type
         - is_current
+    Genome:
+      type: object
+      properties:
+        genome_uuid:
+          type: string
+          format: uuid
+        common_name:
+          type: string
+        scientific_name:
+          type: string
+        assembly_name:
+          type: string
+        is_reference:
+          type: boolean
+        is_group_reference:
+          type: boolean
+        release:
+          $ref: '#/components/schemas/Release'
+    GenomeGroup:
+      type: object
+      properties:
+        group_id:
+          type: string
+        group_type:
+          type: string
+        group_name:
+          type: string
+        reference_genome:
+          $ref: '#/components/schemas/Genome'
   responses:
     404NotFound:
       description: The specified resource was not found
diff --git a/docs/index.html b/docs/index.html
@@ -0,0 +1,30 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <title>API Documentation</title>
+  <link rel="stylesheet" href="https://unpkg.com/swagger-ui-dist/swagger-ui.css" />
+  <style>
+    body { margin: 0; padding: 0; }
+  </style>
+</head>
+<body>
+  <div id="swagger-ui"></div>
+
+  <script src="https://unpkg.com/swagger-ui-dist/swagger-ui-bundle.js"></script>
+  <script src="https://unpkg.com/swagger-ui-dist/swagger-ui-standalone-preset.js"></script>
+  <script>
+    window.onload = () => {
+      SwaggerUIBundle({
+        url: "APISpecification.yaml",  // IMPORTANT: relative path
+        dom_id: '#swagger-ui',
+        presets: [
+          SwaggerUIBundle.presets.apis,
+          SwaggerUIStandalonePreset
+        ],
+        layout: "StandaloneLayout"
+      });
+    };
+  </script>
+</body>
+</html>
