Merge pull request #133 from VariantEffect/bencap/feature/60/expand-genomic-scoreset-tx-pool

Add `cdot` Package to Pyproject.toml for Test Resolution

Author: bencap

pyproject.toml

@@ -25,6 +25,7 @@ dependencies = [
     "eutils~=0.6.0",
     "hgvs~=1.5.4",
     "biocommons~=0.0.0",
+    "cdot~=0.2.21",
     "numpy~=1.22",
     "pandas~=1.4.1",
     "pydantic~=1.10",

src/mavedb/deps.py

@@ -18,14 +18,27 @@ def get_db() -> Generator:
         db.close()
 
 
-def hgvs_data_provider() -> Generator:
-    grch38_fetcher = FastaSeqFetcher("/data/GCF_000001405.39_GRCh38.p13_genomic.fna.gz")
-    grch37_fetcher = FastaSeqFetcher("/data/GCF_000001405.25_GRCh37.p13_genomic.fna.gz")
-
+def hgvs_data_provider() -> RESTDataProvider:
     # Prioritize fetching from SeqRepo, then GRCh38, then GRCh37.
-    seqfetcher = ChainedSeqFetcher(SeqFetcher(), grch38_fetcher, grch37_fetcher)
+    seqfetcher = ChainedSeqFetcher(SeqFetcher())
+
+    # Attempt to resolve FASTA Seq fetchers from data files, but don't fail if neither file is
+    # available. This way, we at least retain some ability to resolve sequences if we don't have
+    # FASTA file access and we are able to run our test suite without needing access to large genomic
+    # files.
+    try:
+        grch38_fetcher = FastaSeqFetcher("/data/GCF_000001405.39_GRCh38.p13_genomic.fna.gz")
+        seqfetcher.seq_fetchers.append(grch38_fetcher)
+    except OSError:
+        pass
+
+    try:
+        grch37_fetcher = FastaSeqFetcher("/data/GCF_000001405.25_GRCh37.p13_genomic.fna.gz")
+        seqfetcher.seq_fetchers.append(grch37_fetcher)
+    except OSError:
+        pass
 
-    yield RESTDataProvider(seqfetcher=seqfetcher)
+    return RESTDataProvider(seqfetcher=seqfetcher)
 
 
 # if 'PYTEST_RUN_CONFIG' in os.environ: