Test SPDI 3 bit packing RefSeq scheme

.github/workflows/main.yml

@@ -28,13 +28,14 @@ jobs:
           args: --extend-ignore E501,E741
 
       - name: Run Tests
-        run: python -m pytest
+        run: ./fetch_refseq.sh && python -m pytest
 
+  # TODOChange this to dev
   deploy:
     name: Deploy
     runs-on: ubuntu-latest
 
-    if: ${{ github.event_name == 'push' && github.ref == 'refs/heads/main' && always() && contains(join(needs.*.result, ','), 'success') }}
+    if: ${{ contains(join(needs.*.result, ','), 'success') }}
     needs: [test]
 
     steps:
@@ -43,5 +44,5 @@ jobs:
       - uses: akhileshns/heroku-deploy@v3.12.12
         with:
           heroku_api_key: ${{secrets.HEROKU_API_KEY}}
-          heroku_app_name: ${{secrets.HEROKU_APP_NAME}}
+          heroku_app_name: ${{secrets.HEROKU_DEV_APP_NAME}}
           heroku_email: ${{secrets.HEROKU_EMAIL}}

README.md

@@ -48,3 +48,9 @@ run `python3 -m pytest` from the terminal to execute them all.
 Additionally, since the tests run against the Mongo DB database, if you need to update the test data in this repo, you
 can run `OVERWRITE_TEST_EXPECTED_DATA=true python3 -m pytest` from the terminal and then create a pull request with the
 changes.
+
+## Development environment on Heroku
+
+Pull requests will trigger a deployment to this environment automatically which is accessible at the following URL:
+
+https://fhir-gen-ops-dev-ca42373833b6.herokuapp.com/

app/__init__.py

@@ -2,13 +2,9 @@
 import flask
 from flask_cors import CORS
 import os
-# from .refseq import download_refseq_files
 
 
 def create_app():
-    # First ensure we have the refseq files locally
-    # download_refseq_files()
-
     # App and API
     options = {
         'swagger_url': '/',

app/api_spec.yml

@@ -102,7 +102,7 @@ paths:
             type: boolean
             default: false
           description: Include sequence phase relationships in response if set to true.
-  
+
   /subject-operations/genotype-operations/$find-subject-specific-variants:
     get:
       description: >
@@ -177,7 +177,7 @@ paths:
               - "germline"
               - "somatic"
           description: Enables an App to limit results to those that are 'germline' or 'somatic'. Default is to include variants irrespective of genomic source class.
-  
+
   /subject-operations/genotype-operations/$find-subject-structural-intersecting-variants:
     get:
       description: >
@@ -262,7 +262,7 @@ paths:
             type: boolean
             default: false
           description: Include variants in response if set to true.
-  
+
   /subject-operations/genotype-operations/$find-subject-structural-subsuming-variants:
     get:
       description: >
@@ -346,7 +346,7 @@ paths:
             type: boolean
             default: false
           description: Include variants in response if set to true.
-  
+
   /subject-operations/genotype-operations/$find-subject-haplotypes:
     get:
       description: >
@@ -422,7 +422,7 @@ paths:
               - "germline"
               - "somatic"
           description: Enables an App to limit results to those that are 'germline' or 'somatic'. Default is to include variants irrespective of genomic source class.
-  
+
   /subject-operations/genotype-operations/$find-subject-specific-haplotypes:
     get:
       description: >
@@ -497,7 +497,7 @@ paths:
               - "germline"
               - "somatic"
           description: Enables an App to limit results to those that are 'germline' or 'somatic'. Default is to include variants irrespective of genomic source class.
-  
+
   /subject-operations/phenotype-operations/$find-subject-tx-implications:
     get:
       description: |-
@@ -614,7 +614,7 @@ paths:
               - "germline"
               - "somatic"
           description: Enables an App to limit results to those that are 'germline' or 'somatic'. Default is to include variants irrespective of genomic source class.
-  
+
   /subject-operations/phenotype-operations/$find-subject-dx-implications:
     get:
       description: |-
@@ -713,7 +713,7 @@ paths:
               - "germline"
               - "somatic"
           description: Enables an App to limit results to those that are 'germline' or 'somatic'. Default is to include variants irrespective of genomic source class.
-  
+
   /subject-operations/metadata-operations/$find-study-metadata:
     get:
       description: |-
@@ -1147,6 +1147,7 @@ paths:
             type: string
             pattern: '^\s*[Nn][Pp]_\d{4,10}(\.)?(\d{1,2})?\s*$'
           example: "NP_000005.3"
+
   /utilities/find-the-gene:
     get:
       summary: "Find The Gene"
@@ -1170,6 +1171,42 @@ paths:
             pattern: '^\s*[Nn][Cc]_\d{4,10}(\.)(\d{1,2}):\d{1,10}-\d{1,10}\s*$'
             example: "NC_000001.11:11794399-11794400"
 
+  /utilities/seqfetcher/1/sequence/{ref_seq}:
+    get:
+      summary: "Seqfetcher"
+      operationId: "app.utilities_endpoints.seqfetcher"
+      tags:
+        - "Seqfetcher Utility"
+      responses:
+        '200':
+          description: "Returns RefSeq subsequence"
+          content:
+            text/plain:
+              schema:
+                type: string
+      parameters:
+        - name: ref_seq
+          in: path
+          required: true
+          description: RefSeq
+          schema:
+            type: string
+            example: "NC_000001.10"
+        - name: start
+          in: query
+          required: true
+          description: Subsequence start index
+          schema:
+            type: integer
+            example: 1
+        - name: end
+          in: query
+          required: true
+          description: Subsequence end index
+          schema:
+            type: integer
+            example: 2
+
 tags:
   - name: Subject Genotype Operations
   - name: Subject Phenotype Operations
@@ -1178,7 +1215,7 @@ tags:
   - name: Population Phenotype Operations
   - name: Feature Coordinates Utility
     description: This utility returns genomic feature coordinates and other annotations. All data are from <a href="https://www.ncbi.nlm.nih.gov/genome/guide/human/">NCBI Human Genome Resources</a>. For chromosomes, build 37 and build 38 reference sequences are returned. For genes, genomic coordinates are returned, along with a list of transcripts. MANE transcript is flagged. For transcripts, genomic coordinates are returned, along with the gene name and composite exons, along with exon coordinates. For proteins, the corresponding transcript is returned.
-  
+
   - name: Find The Gene Utility
     description: This utility returns all genes that intersect with a provided genomic region.
 

app/utilities_endpoints.py

@@ -1,6 +1,7 @@
 from flask import abort, jsonify
 from collections import OrderedDict
 from app import common
+from utilities import SPDI_Normalization
 
 
 def get_feature_coordinates(
@@ -133,7 +134,8 @@ def get_feature_coordinates(
             protein = protein.split('.')[0]
 
         try:
-            result = common.proteins_data.aggregate([{"$match": {"proteinRefSeq": {'$regex': ".*"+str(protein).replace('*', r'\*')+".*"}}}])
+            result = common.proteins_data.aggregate(
+                [{"$match": {"proteinRefSeq": {'$regex': ".*"+str(protein).replace('*', r'\*')+".*"}}}])
             result = list(result)
         except Exception as e:
             print(f"DEBUG: Error({e}) under get_feature_coordinates(protein={protein})")
@@ -189,3 +191,11 @@ def find_the_gene(range=None):
         output.append(ord_dict)
 
     return (jsonify(output))
+
+
+def seqfetcher(ref_seq, start, end):
+    try:
+        subseq = SPDI_Normalization.get_ref_seq_subseq('GRCh37', ref_seq, start, end)
+    except Exception:
+        subseq = SPDI_Normalization.get_ref_seq_subseq('GRCh38', ref_seq, start, end)
+    return f'>{ref_seq}:{start}-{end} Homo sapiens chromosome 1, GRCh37.p13 Primary Assembly\n{subseq}\n\n'

fetch_refseq.sh

@@ -10,12 +10,12 @@ cd ./refseq
 
 echo "Downloading refseq files..."
 
-curl -sLO https://github.com/FHIR/genomics-operations/releases/download/113c119/GRCh37seq.tar.gz
-curl -sLO https://github.com/FHIR/genomics-operations/releases/download/113c119/GRCh38seq.tar.gz
+curl -sLO https://github.com/FHIR/genomics-operations/releases/download/113c119/GRCh37_refseq.tar.gz
+curl -sLO https://github.com/FHIR/genomics-operations/releases/download/113c119/GRCh38_refseq.tar.gz
 
 echo "Extracting refseq files..."
 
-tar -xzf GRCh37seq.tar.gz
-tar -xzf GRCh38seq.tar.gz
+tar -xzf GRCh37_refseq.tar.gz
+tar -xzf GRCh38_refseq.tar.gz
 
 echo "Finished extracting refseq files."