Add initial Dmel and GH actions

.github/workflows/generate-blast-conf.yml

@@ -0,0 +1,59 @@
+name: Generate FlyBase BLAST Configuration
+on:
+  workflow_dispatch:
+    inputs:
+      FB-release:
+        required: true
+      dmel-annot-release:
+        required: true
+
+jobs:
+  generate:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository code
+        uses: actions/checkout@v4
+
+      - name: Set up python
+        id: setup-python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+
+      - name: Install Poetry
+        uses: snok/install-poetry@v1
+        with:
+          virtualenvs-create: true
+          virtualenvs-in-project: true
+          installer-parallel: true
+
+      #----------------------------------------------
+      #       load cached venv if cache exists
+      #----------------------------------------------
+      - name: Load cached venv
+        id: cached-poetry-dependencies
+        uses: actions/cache@v3
+        with:
+          path: .venv
+          key: venv-${{ runner.os }}-${{ steps.setup-python.outputs.python-version }}-${{ hashFiles('**/poetry.lock') }}
+
+      #----------------------------------------------
+      # install dependencies if cache does not exist
+      #----------------------------------------------
+      - name: Install dependencies
+        if: steps.cached-poetry-dependencies.outputs.cache-hit != 'true'
+        run: poetry install --no-interaction --no-root
+
+      - name: Generate configuration
+        run: poetry run generate --release ${{ github.event.inputs.FB-release }} --dmel-annot-release ${{ github.event.inputs.dmel-annot-release }}
+
+      - name: Create PR
+        uses: peter-evans/create-pull-request@v6
+        with:
+          commit-message: ${{ github.event.inputs.FB-release }} BLAST config
+          branch: ${{ github.event.inputs.FB-release }}-blast-config
+          delete-branch: true
+          title: '[Update] ${{ github.event.inputs.FB-release }} BLAST config'
+          body: |
+            ${{ github.event.inputs.FB-release }} BLAST config
+          draft: false

.pre-commit-config.yaml

@@ -0,0 +1,27 @@
+# .pre-commit-config.yaml
+# See https://pre-commit.com for more information
+# See https://pre-commit.com/hooks.html for more hooks
+repos:
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v4.5.0
+    hooks:
+      - id: check-case-conflict
+      - id: check-merge-conflict
+      - id: trailing-whitespace
+      - id: end-of-file-fixer
+      - id: check-yaml
+      - id: check-toml
+      - id: check-json
+      - id: check-added-large-files
+  - repo: local
+    hooks:
+      - id: black
+        name: black
+        entry: poetry run black
+        language: system
+        types: [file, python]
+      - id: isort
+        name: isort
+        entry: poetry run isort
+        language: system
+        types: [file, python]

pyproject.toml

@@ -19,6 +19,7 @@ tqdm = "^4.66.4"
 
 [tool.poetry.group.dev.dependencies]
 black = "^24.4.2"
+isort = "^5.13.2"
 
 [build-system]
 requires = ["poetry-core"]

src/blast_db_configuration/__main__.py

@@ -1,15 +1,16 @@
+import json
+import logging
 from dataclasses import dataclass
-from typing_extensions import Annotated
 from datetime import datetime
 from pathlib import Path
-import json
-import logging
+
+import agr_blast_service_configuration.schemas.metadata as agrdb
 import typer
-from tqdm import tqdm
 from Bio import Entrez
-import agr_blast_service_configuration.schemas.metadata as agrdb
+from tqdm import tqdm
+from typing_extensions import Annotated
 
-from .db_metadata import create_metadata_from_ncbi
+from .db_metadata import create_dmel_metadata, create_metadata_from_ncbi
 
 app = typer.Typer()
 
@@ -44,6 +45,9 @@ class DefaultBlastDbConfiguration:
 @app.command()
 def generate_config(
     release: Annotated[str, typer.Option(help="The FlyBase release version")],
+    dmel_annot_release: Annotated[
+        str, typer.Option(help="The Dmel annotation release version e.g. r6.57")
+    ],
     contact: Annotated[
         type(DEFAULT_CONFIG.contact),
         typer.Option(help="Email of the FlyBase technical contact."),
@@ -99,7 +103,7 @@ def generate_config(
         unit="organism",
     ):
         if genus == "Drosophila" and species == "melanogaster":
-            pass
+            all_dbs.extend(create_dmel_metadata(dmel_annot_release))
         else:
             all_dbs.extend(create_metadata_from_ncbi(genus, species, ncbi_email))
 

src/blast_db_configuration/db_metadata.py

@@ -1,9 +1,11 @@
 import logging
+import urllib.request
+from typing import Optional
 
 import agr_blast_service_configuration.schemas.metadata as blast_metadata_schema
 
-from .ncbi import taxonomy as tax
 from .ncbi import genomes as genomes
+from .ncbi import taxonomy as tax
 
 logger = logging.getLogger(__name__)
 
@@ -72,39 +74,72 @@ def create_metadata_from_ncbi(
     return dbs
 
 
-def create_dmel_metadata():
-    pass
-    # dbs.extend(
-    #     [
-    #         blast_metadata.BlastDBMetaData(
-    #             version=options.dmel_annot,
-    #             URI=f"ftp://ftp.flybase.org/genomes/Drosophila_melanogaster/dmel_r{options.dmel_annot}_{options.release}/fasta/dmel-all-chromosome-r{options.dmel_annot}.fasta.gz",
-    #             md5sum="b7bc17acfd655914c68326df8599a9ca",  # TODO - Hard coded for now, need to fetch this from the MD5SUM file
-    #             genus="Drosophila",
-    #             species="melanogaster",
-    #             blast_title=f"D. melanogaster Genome Assembly ({options.dmel_annot})",
-    #             description="Drosophila melanogaster genome assembly",
-    #             taxon_id="NCBITaxon:7227",
-    #             seqtype="nucl",
-    #         ),
-    #         blast_metadata.BlastDBMetaData(
-    #             version=options.dmel_annot,
-    #             URI=f"ftp://ftp.flybase.org/genomes/Drosophila_melanogaster/dmel_r{options.dmel_annot}_{options.release}/fasta/dmel-all-translation-r{options.dmel_annot}.fasta.gz",
-    #             # TODO - Hard coded for now, need to fetch this from the MD5SUM file
-    #             md5sum="e3f959ab0e1026de56e1bd00490450e5",
-    #             genus="Drosophila",
-    #             species="melanogaster",
-    #             blast_title=f"D. melanogaster Protein Sequences ({options.dmel_annot})",
-    #             description="Drosophila melanogaster protein sequences",
-    #             taxon_id="NCBITaxon:7227",
-    #             seqtype="prot",
-    #         ),
-    #     ]
-    # )
-    # flybase_blast_metadata = blast_metadata.AGRBlastDatabases(
-    #     metaData=blast_metadata.AGRBlastMetadata(
-    #         contact=options.email, dataProvider="FlyBase", release=options.release
-    #     ),
-    #     data=dbs,
-    # )
-    # print(flybase_blast_metadata.json())
+def create_dmel_metadata(
+    dmel_annot_release: str,
+) -> list[blast_metadata_schema.SequenceMetadata]:
+    """
+    Generate a list of BLAST DB metadata schemas based on Dmel annot release.
+
+    :param dmel_annot_release: The Dmel annot release
+    :return: List of BLAST DB metadata schemas
+    """
+    dmel_dbs = [
+        {
+            "uri": "https://ftp.flybase.org/blast/dmel-assembly.fasta.gz",
+            "description": f"D. melanogaster Genome Assembly {dmel_annot_release}",
+            "seqtype": blast_metadata_schema.BlastDBType.NUCL,
+            "md5_sum": None,
+        },
+        {
+            "uri": "https://ftp.flybase.org/blast/dmel-intergenic.fasta.gz",
+            "description": f"D. melanogaster Intergenic Regions {dmel_annot_release}",
+            "seqtype": blast_metadata_schema.BlastDBType.NUCL,
+            "md5_sum": None,
+        },
+        {
+            "uri": "https://ftp.flybase.org/blast/dmel-transcript.fasta.gz",
+            "description": f"D. melanogaster Transcripts {dmel_annot_release}",
+            "seqtype": blast_metadata_schema.BlastDBType.NUCL,
+            "md5_sum": None,
+        },
+        {
+            "uri": "https://ftp.flybase.org/blast/dmel-translation.fasta.gz",
+            "description": f"D. melanogaster Proteins {dmel_annot_release}",
+            "seqtype": blast_metadata_schema.BlastDBType.PROT,
+            "md5_sum": None,
+        },
+        {
+            "uri": "https://ftp.flybase.org/blast/dmel-transposon.fasta.gz",
+            "description": f"D. melanogaster Transposons {dmel_annot_release}",
+            "seqtype": blast_metadata_schema.BlastDBType.NUCL,
+            "md5_sum": None,
+        },
+    ]
+    # TODO: read in checksums and assign them to the appropriate DB.
+    return [
+        blast_metadata_schema.SequenceMetadata(
+            version=dmel_annot_release,
+            uri=db.get("uri"),
+            md5_sum="MD5",
+            genus="Drosophila",
+            species="melanogaster",
+            blast_title=db.get("description"),
+            description=db.get("description"),
+            taxon_id="7227",
+            seqtype=db.get("seqtype"),
+        )
+        for db in dmel_dbs
+    ]
+
+
+def fetch_dmel_checksums(uri: str) -> Optional[str]:
+    """
+    Get the current Dmel FASTA checksums.
+
+    :param uri: The URI of the checksum file
+    :return: The text content of the checksum file
+    """
+    with urllib.request.urlopen(uri) as response:
+        md5_checksums = response.read().decode("utf-8")
+        return md5_checksums
+    return None