feat: Adding Clinvar Accession to TSV input (#146)

clinvar_this/batches.py

@@ -88,8 +88,10 @@ def merge_submission(
         base: models.SubmissionClinvarSubmission,
         patch: models.SubmissionClinvarSubmission,
     ) -> models.SubmissionClinvarSubmission:
+        clinvar_accession = base.clinvar_accession or patch.clinvar_accession
         return evolve(
             base,
+            clinvar_accession=clinvar_accession,
             condition_set=patch.condition_set,
             clinical_significance=patch.clinical_significance,
             observed_in=patch.observed_in,
@@ -248,7 +250,7 @@ def _retrieve_store_response(
         submissions = summary_response.submissions or []
         for submission in submissions:
             local_key_to_accession[
-                submission.identifiers.local_key
+                submission.identifiers.local_key or submission.identifiers.clinvar_local_key
             ] = submission.identifiers.clinvar_accession
             errors = [
                 error_inner.user_message

clinvar_this/io/tsv.py

@@ -74,6 +74,8 @@ class SeqVarTsvRecord:
     clinical_significance_comment: typing.Optional[str] = None
     #: HPO terms for clinical features
     hpo_terms: typing.Optional[typing.List[str]] = None
+    #: Existing ClinVar SCV accession
+    accession: typing.Optional[str] = None
 
 
 @attrs.frozen
@@ -106,6 +108,8 @@ class StrucVarTsvRecord:
     clinical_significance_comment: typing.Optional[str] = None
     #: HPO terms for clinical features
     hpo_terms: typing.Optional[typing.List[str]] = None
+    #: Existing ClinVar SCV accession
+    accession: typing.Optional[str] = None
 
 
 @attrs.frozen
@@ -275,6 +279,13 @@ def _join_list(xs: typing.List[typing.Any]) -> str:
         converter=_str_list,
         extractor=lambda r: _join_list(r.hpo_terms or []),
     ),
+    SeqVarHeaderColumn(
+        header_names=("ACCESSION",),
+        key="accession",
+        required=False,
+        converter=str,
+        extractor=lambda r: str(r.accession or ""),
+    ),
 )
 
 #: The header columns for structural variant TSV files.
@@ -363,6 +374,13 @@ def _join_list(xs: typing.List[typing.Any]) -> str:
         converter=_str_list,
         extractor=lambda r: _join_list(r.omim),
     ),
+    StrucVarHeaderColumn(
+        header_names=("ACCESSION",),
+        key="accession",
+        required=False,
+        converter=str,
+        extractor=lambda r: str(r.accession or ""),
+    ),
 )
 
 
@@ -695,6 +713,7 @@ def record_clinical_features(
         clinvar_submission_release_status=release_status,
         clinvar_submission=[
             SubmissionClinvarSubmission(
+                clinvar_accession=record.accession,
                 local_id=str(_uuid4_if_falsy()),
                 local_key=record.local_key,
                 condition_set=SubmissionConditionSet(condition=[record_condition(record)]),
@@ -778,6 +797,7 @@ def record_clinical_features(
         clinvar_submission_release_status=release_status,
         clinvar_submission=[
             SubmissionClinvarSubmission(
+                clinvar_accession=record.accession,
                 local_id=str(_uuid4_if_falsy()),
                 local_key=record.local_key,
                 condition_set=SubmissionConditionSet(condition=[record_condition(record)]),
@@ -883,12 +903,11 @@ def submission_to_seq_var_tsv_record(
             )
 
         extra_data = {}
-        if submission.clinvar_accession:
-            extra_data["clinvar_accession"] = submission.clinvar_accession  # XXX
         if submission.extra_data:
             extra_data.update(submission.extra_data)
 
         return SeqVarTsvRecord(
+            accession=submission.clinvar_accession,
             assembly=chromosome_coordinates.assembly,
             chromosome=chromosome_coordinates.chromosome,
             pos=chromosome_coordinates.start,
@@ -980,12 +999,11 @@ def submission_to_struc_var_tsv_record(
             )
 
         extra_data = {}
-        if submission.clinvar_accession:
-            extra_data["clinvar_accession"] = submission.clinvar_accession  # XXX
         if submission.extra_data:
             extra_data.update(submission.extra_data)
 
         return StrucVarTsvRecord(
+            accession=submission.clinvar_accession,
             assembly=chromosome_coordinates.assembly,
             chromosome=chromosome_coordinates.chromosome,
             start=chromosome_coordinates.start,

tests/clinvar_this/data/io_tsv/example.tsv

@@ -1,2 +1,2 @@
-ASSEMBLY	CHROM	POS	REF	ALT	OMIM	MOI	CLIN_SIG	KEY	gene
-GRCh37	10	115614632	A	G	OMIM:618278	Autosomal recessive inheritance	not provided	KEY	NHLRC2
+ASSEMBLY	CHROM	POS	REF	ALT	OMIM	MOI	CLIN_SIG	KEY	gene	ACCESSION
+GRCh37	10	115614632	A	G	OMIM:618278	Autosomal recessive inheritance	not provided	KEY	NHLRC2	SCV1

tests/clinvar_this/data/io_tsv/example_sv.tsv

@@ -1,2 +1,2 @@
-ASSEMBLY	CHROM	START	STOP	SV_TYPE	OMIM	MOI	CLIN_SIG	HPO
-GRCh38	1	844347	4398122	Deletion		Autosomal dominant inheritance	not provided	HP:0001263
+ASSEMBLY	CHROM	START	STOP	SV_TYPE	OMIM	MOI	CLIN_SIG	HPO	ACCESSION
+GRCh38	1	844347	4398122	Deletion		Autosomal dominant inheritance	not provided	HP:0001263	SCV1

tests/clinvar_this/test_batches.py

@@ -268,10 +268,10 @@ def test_export_seq_variant_tsv(fs, app_config, exists, force):
     if not exists or force:
         expected = "\n".join(
             [
-                "ASSEMBLY\tCHROM\tPOS\tREF\tALT\tOMIM\tMOI\tCLIN_SIG\tCLIN_EVAL\tCLIN_COMMENT\tKEY\tHPO",
+                "ASSEMBLY\tCHROM\tPOS\tREF\tALT\tOMIM\tMOI\tCLIN_SIG\tCLIN_EVAL\tCLIN_COMMENT\tKEY\tHPO\tACCESSION",
                 (
                     "GRCh37\t19\t48183936\tC\tCA\t619325\tAutosomal dominant inheritance\t"
-                    "Likely pathogenic\t\t\t\tHP:0004322,HP:0001263\n"
+                    "Likely pathogenic\t\t\t\tHP:0004322,HP:0001263\t\n"
                 ),
             ]
         )
@@ -319,11 +319,11 @@ def test_export_structural_variant_tsv(fs, app_config, exists, force):
             [
                 (
                     "ASSEMBLY\tCHROM\tSTART\tSTOP\tSV_TYPE\tOMIM\tMOI\tCLIN_SIG\tCLIN_EVAL\t"
-                    "CLIN_COMMENT\tKEY\tHPO"
+                    "CLIN_COMMENT\tKEY\tHPO\tACCESSION"
                 ),
                 (
                     "GRCh38\t1\t844347\t4398122\tDeletion\t\tAutosomal dominant inheritance\t"
-                    "not provided\t\t\t\t\n"
+                    "not provided\t\t\t\t\t\n"
                 ),
             ]
         )

tests/clinvar_this/test_io_tsv.py

@@ -26,6 +26,7 @@ def test_read_seq_var_tsv_path():
     actual = read_seq_var_tsv(path=DATA_DIR / "example.tsv")
     assert actual == [
         SeqVarTsvRecord(
+            accession="SCV1",
             assembly=Assembly.GRCH37,
             chromosome=Chromosome.CHR10,
             pos=115614632,
@@ -44,6 +45,7 @@ def test_read_struc_var_tsv_path():
     actual = read_struc_var_tsv(path=DATA_DIR / "example_sv.tsv")
     assert actual == [
         StrucVarTsvRecord(
+            accession="SCV1",
             assembly=Assembly.GRCH38,
             chromosome=Chromosome.CHR1,
             start=844347,
@@ -62,6 +64,7 @@ def test_read_seq_var_tsv_file():
         actual = read_seq_var_tsv(file=inputf)
     assert actual == [
         SeqVarTsvRecord(
+            accession="SCV1",
             assembly=Assembly.GRCH37,
             chromosome=Chromosome.CHR10,
             pos=115614632,
@@ -81,6 +84,7 @@ def test_read_struc_var_tsv_file():
         actual = read_struc_var_tsv(file=inputf)
     assert actual == [
         StrucVarTsvRecord(
+            accession="SCV1",
             assembly=Assembly.GRCH38,
             chromosome=Chromosome.CHR1,
             start=844347,