Merge pull request #319 from NCI-CGR/issue-210-revisit-disc-exp-reps

fix: Handle removal of discordant expected duplicates (issue #210)
NCI-CGR · Sep 6, 2024 · 0987eb1 · 0987eb1
2 parents 5cf67de + 3eff185
commit 0987eb1
Show file tree

Hide file tree

Showing 4 changed files with 10 additions and 10 deletions.
diff --git a/src/cgr_gwas_qc/workflow/conda/illuminaio.yml b/src/cgr_gwas_qc/workflow/conda/illuminaio.yml
@@ -2,4 +2,4 @@ channels:
   - bioconda
   - conda-forge
 dependencies:
-  - bioconductor-illuminaio=0.20.0
+  - bioconductor-illuminaio=0.44.0
diff --git a/src/cgr_gwas_qc/workflow/scripts/sample_qc_table.py b/src/cgr_gwas_qc/workflow/scripts/sample_qc_table.py
@@ -194,18 +194,14 @@ def main(
     add_qc_columns(
         sample_qc, remove_contam, remove_rep_discordant,
     )
-    sample_qc["is_unexpected_replicate"] = (
-        sample_qc["is_unexpected_replicate"].replace("", False).fillna(False)
-    )
-    sample_qc["is_discordant_replicate"] = (
-        sample_qc["is_discordant_replicate"].replace("", False).fillna(False)
-    )
+
     sample_qc = sample_qc.rename(
         columns={
             "is_unexpected_replicate": "Unexpected Replicate",
             "is_discordant_replicate": "Expected Replicate Discordance",
         }
     )
+
     save(sample_qc, outfile)
 
 
@@ -396,6 +392,8 @@ def _read_concordance(filename: Path, Sample_IDs: pd.Index) -> pd.DataFrame:
         .max()  # Flag a sample as True if it is True for any comparison.
         .astype("boolean")
         .reindex(Sample_IDs)
+        .replace("", False)
+        .fillna(False)
     )
 
 
@@ -527,7 +525,7 @@ def _add_analytic_exclusion(
         exclusion_criteria["is_contaminated"] = "Contamination"
 
     if remove_rep_discordant:
-        exclusion_criteria["Expected Replicate Discordance"] = "Replicate Discordance"
+        exclusion_criteria["is_discordant_replicate"] = "Replicate Discordance"
 
     sample_qc["analytic_exclusion"] = sample_qc.reindex(exclusion_criteria.keys(), axis=1).any(
         axis=1

diff --git a/src/cgr_gwas_qc/workflow/sub_workflows/entry_points.smk b/src/cgr_gwas_qc/workflow/sub_workflows/entry_points.smk
@@ -88,13 +88,15 @@ if cfg.config.user_files.gtc_pattern:
             cfg.conda("bcftools-gtc2vcf-plugin")
         shell:
             "bcftools +gtc2vcf --gtcs {input.gtcs} --bpm {input.bpm} --fasta-ref {input.reference_fasta} --output {output.vcf} --use-gtc-sample-names"
+
     rule filter_missing_allele_snps:
         input:
             vcf=rules.gtc_to_vcf.output.vcf,
         output:
             vcf=temp("sample_level/samples_filtered.vcf"),
         shell:
-            "grep -vP '\t\.\t\.\t\.' {input.vcf} > {output.vcf}" 
+            "grep -vP '\t\.\t\.\t\.' {input.vcf} > {output.vcf}"
+
     rule vcf_to_bed:
         input:
             vcf=rules.filter_missing_allele_snps.output.vcf,

diff --git a/tests/workflow/scripts/test_sample_qc_table.py b/tests/workflow/scripts/test_sample_qc_table.py
@@ -254,7 +254,7 @@ def fake_sample_qc() -> pd.DataFrame:
         "is_cr1_filtered",
         "is_cr2_filtered",
         "is_contaminated",
-        "Expected Replicate Discordance",
+        "is_discordant_replicate",
     ]
     data = [
         ("SP00001", "SB00001", False, False, 0.99, False, False, False, False),