Merge pull request #290 from NCI-CGR/281-ancestry-=-unknown-not-shown…

…-in-table-3-subject-counts-by-ancestry-group

281 ancestry = unknown not shown in table 3 subject counts by ancestry group

src/cgr_gwas_qc/reporting/sample_qc.py

@@ -27,7 +27,6 @@ def construct(
         call_rate_png: Path,
         ancestry_png: Path,
     ) -> "SampleQC":
-
         return cls(
             ArrayProcessing.construct(sample_sheet),
             CompletionRate.construct(snp_qc, sample_qc, call_rate_png),
@@ -206,7 +205,6 @@ def _count_subjects(sample_qc: pd.DataFrame) -> int:
 
 @dataclass
 class Ancestry:
-
     table: str
     png: str
 
@@ -216,6 +214,12 @@ def construct(cls, sample_qc: pd.DataFrame, png: Path) -> "Ancestry":
 
     @staticmethod
     def _build_table(sample_qc: pd.DataFrame) -> str:
+        # Issue 281: Ensure 'Other' ancestry output from graf-pop captured in table 3
+
+        # Add 'Other' to the categories and full blanks with "Other"
+        sample_qc["Ancestry"] = sample_qc["Ancestry"].cat.add_categories(["Other"])
+        sample_qc["Ancestry"] = sample_qc["Ancestry"].fillna("Other")
+
         return (
             sample_qc.query("is_subject_representative")
             .groupby("Ancestry")