WIP

clinvar_this/io/tsv.py

@@ -109,6 +109,8 @@ class StrucVarTsvRecord:
     clinical_significance_comment: typing.Optional[str] = None
     #: HPO terms for clinical features
     hpo_terms: typing.Optional[typing.List[str]] = None
+    #: Pubmed PMID for literature references
+    pmids: typing.Optional[typing.List[str]] = None
 
 
 @attrs.frozen
@@ -373,6 +375,13 @@ def _join_list(xs: typing.List[typing.Any]) -> str:
         converter=_str_list,
         extractor=lambda r: _join_list(r.omim),
     ),
+    StrucVarHeaderColumn(
+        header_names=("PMID",),
+        key="hpo_terms",
+        required=False,
+        converter=_str_list,
+        extractor=lambda r: _join_list(r.pmids or []),
+    ),
 )
 
 
@@ -965,6 +974,10 @@ def _inheritance(submission: SubmissionClinvarSubmission) -> typing.Optional[Mod
         else:
             return None
 
+    def _pmids(submission: SubmissionClinvarSubmission) -> typing.Optional[typing.List[str]]:
+        if citations := submission.clinical_significance.citation:
+            return [c.id for c in citations if c.db == CitationDb.PUBMED]
+
     def submission_to_struc_var_tsv_record(
         submission: SubmissionClinvarSubmission,
     ) -> typing.Optional[StrucVarTsvRecord]:
@@ -1020,6 +1033,7 @@ def submission_to_struc_var_tsv_record(
             stop=chromosome_coordinates.stop,
             sv_type=variant_type,
             omim=_condition(submission),
+            pmids=_pmids(submission),
             inheritance=_inheritance(submission),
             clinical_significance_description=submission.clinical_significance.clinical_significance_description,
             local_key=submission.local_key or "",

docs/file_formats.rst

@@ -37,6 +37,9 @@ The following headers are optional:
   Filled automatically with a UUID if missing, recommeded to leave empty.
 - ``HPO`` - List of HPO terms separated by comma or semicolon, any space will be stripped.
   E.g., ``HP:0004322; HP:0001263``.
+- ``PMID`` - List of Pubmed IDs separated by a comma or semicolon, any space
+  will be stripped.
+  E.g., ``31859447‚29474920``.
 - ``$remove_from_batch`` - you can use this for removing a previously added variant from the given batch; one of ``true`` and ``false``, defaults to ``false``.
 
 Any further header will be imported into the local repository into an ``extra_data`` field.
@@ -88,6 +91,9 @@ The following headers are optional:
   Filled automatically with a UUID if missing, recommeded to leave empty.
 - ``HPO`` - List of HPO terms separated by comma or semicolon, any space will be stripped.
   E.g., ``HP:0004322; HP:0001263``.
+- ``PMID`` - List of Pubmed IDs separated by a comma or semicolon, any space
+  will be stripped.
+  E.g., ``31859447‚29474920``.
 - ``$remove_from_batch`` - you can use this for removing a previously added variant from the given batch; one of ``true`` and ``false``, defaults to ``false``.
 
 Any further header will be imported into the local repository into an ``extra_data`` field.

tests/clinvar_this/data/io_tsv/example_citation.tsv

@@ -0,0 +1,2 @@
+ASSEMBLY	CHROM	POS	REF	ALT	OMIM	MOI	CLIN_SIG	KEY	gene	PMID
+GRCh37	10	115614632	A	G	OMIM:618278	Autosomal recessive inheritance	not provided	KEY	NHLRC2	123456,000001

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\tPMID",
                 (
                     "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\tPMID"
                 ),
                 (
                     "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

@@ -76,6 +76,26 @@ def test_read_seq_var_tsv_file():
     ]
 
 
+def test_read_seq_var_tsv_file_citation():
+    with (DATA_DIR / "example_citation.tsv").open("rt") as inputf:
+        actual = read_seq_var_tsv(file=inputf)
+    assert actual == [
+        SeqVarTsvRecord(
+            assembly=Assembly.GRCH37,
+            chromosome=Chromosome.CHR10,
+            pos=115614632,
+            ref="A",
+            alt="G",
+            omim=["OMIM:618278"],
+            inheritance=ModeOfInheritance.AUTOSOMAL_RECESSIVE_INHERITANCE,
+            clinical_significance_description=ClinicalSignificanceDescription.NOT_PROVIDED,
+            local_key="KEY",
+            pmids=["123456", "000001"],
+            extra_data={"gene": "NHLRC2"},
+        )
+    ]
+
+
 def test_read_struc_var_tsv_file():
     with (DATA_DIR / "example_sv.tsv").open("rt") as inputf:
         actual = read_struc_var_tsv(file=inputf)