feat: extract additional WGS metrics

BALSAMIC/assets/scripts/collect_qc_metrics.py

@@ -49,6 +49,7 @@ def collect_qc_metrics(
 
 def capture_kit_resolve_type(capture_kit: str):
     """Resolves the capture_kit type (NoneType or String)"""
+
     if capture_kit == "None":
         return None
     else:
@@ -67,9 +68,12 @@ def get_multiqc_data_source(multiqc_data: dict, sample: str, tool: str) -> str:
         A source file that was used to produce a specific metric
     """
 
-    # Use case: splits multiqc_picard_dups into ['multiqc', 'picard', 'dup'] in order to retrieve the
-    # ["report_data_sources"]["Picard"]["DuplicationMetrics"] values from multiqc_data.json
-    subtool_name = tool[:-1].split("_")
+    if tool == "multiqc_general_stats":
+        subtool_name = ["multiqc", "FastQC", "all_sections"]
+    else:
+        # Use case: splits multiqc_picard_dups into ['multiqc', 'picard', 'dup'] in order to retrieve the
+        # ["report_data_sources"]["Picard"]["DuplicationMetrics"] values from multiqc_data.json
+        subtool_name = tool[:-1].split("_")
 
     # Nested json fetching
     for source_tool in multiqc_data["report_data_sources"]:
@@ -139,6 +143,7 @@ def extract(data, output_metrics, sample=None, source=None):
 
         if isinstance(data, dict):
             for k in data:
+                # Ignore UMI and reverse reads metrics
                 if "umi" not in k:
                     if k in requested_metrics:
                         output_metrics.append(

BALSAMIC/constants/quality_check_reporting.py

@@ -90,5 +90,14 @@
             "FOLD_80_BASE_PENALTY": {"condition": {"norm": "lt", "threshold": 1.8}},
         },
     },
-    "wgs": {"FOLD_80_BASE_PENALTY": {"condition": {"norm": "lt", "threshold": 1.8}}},
+    "wgs": {
+        "MEAN_INSERT_SIZE": {"condition": None},
+        "MEDIAN_COVERAGE": {"condition": None},
+        "FastQC_mqc-generalstats-fastqc-percent_duplicates": {"condition": None},
+        "PCT_15X": {"condition": None},
+        "PCT_30X": {"condition": None},
+        "PCT_60X": {"condition": None},
+        "PCT_100X": {"condition": None},
+        "FOLD_80_BASE_PENALTY": {"condition": {"norm": "lt", "threshold": 1.8}},
+    },
 }

BALSAMIC/utils/models.py

@@ -739,6 +739,15 @@ class MetricModel(BaseModel):
     value: Any = ...
     condition: Optional[MetricConditionModel] = ...
 
+    @validator("name")
+    def validate_name(cls, name, values):
+        """Updates the name if the source is FastQC"""
+
+        if "fastqc-percent_duplicates" in name:
+            return "PERCENT_DUPLICATION_R" + values["input"].split("_")[-2]
+
+        return name
+
 
 class MetricValidationModel(BaseModel):
     """Defines the metric validation model
@@ -753,7 +762,7 @@ class MetricValidationModel(BaseModel):
     metrics: List[MetricModel]
 
     @validator("metrics", each_item=True)
-    def check_squares(cls, metric):
+    def validate_metrics(cls, metric):
         """Checks if a metric meets its filtering condition"""
 
         if metric.condition and not VALID_OPS[metric.condition.norm](

CHANGELOG.rst

@@ -17,6 +17,7 @@ Added:
 * Snakemake rule for creating `.cgh` files from `CNVkit` outputs #858 
 * SVdb to TGA workflow #871 
 * SVdb merge SV and CNV #871
+* Additional WGS metrics to be stored in ``<case>_metrics_deliverables.yaml`` #907
 
 Changed:
 ^^^^^^^^

tests/test_data/qc_files/analysis/qc/multiqc_data/multiqc_data.json

@@ -11,6 +11,14 @@
             "DuplicationMetrics": {
                 "concatenated_tumor_XXXXXX_R": "tests/test_data/qc_files/analysis/bam/concatenated_tumor_XXXXXX_R.sorted.mrkdup.txt"
             }
+        },
+        "FastQC": {
+            "all_sections": {
+                "concatenated_tumor_XXXXXX_R_2": "tests/test_data/qc_files/analysis/fastqc/concatenated_tumor_XXXXXX_R_2_fastqc.zip",
+                "concatenated_normal_XXXXXX_R_1": "tests/test_data/qc_files/analysis/fastqc/concatenated_normal_XXXXXX_R_1_fastqc.zip",
+                "concatenated_normal_XXXXXX_R_2": "tests/test_data/qc_files/analysis/fastqc/concatenated_normal_XXXXXX_R_2_fastqc.zip",
+                "concatenated_tumor_XXXXXX_R_1": "tests/test_data/qc_files/analysis/fastqc/concatenated_tumor_XXXXXX_R_1_fastqc.zip"
+            }
         }
     },
     "report_saved_raw_data": {
@@ -76,6 +84,24 @@
                 "READ_PAIR_DUPLICATES": 18741892.0,
                 "PERCENT_DUPLICATION": 0.391429
             }
+        },
+        "multiqc_general_stats": {
+            "concatenated_tumor_XXXXXX_R_2": {
+                "FastQC_mqc-generalstats-fastqc-percent_duplicates": 15.03521942842923,
+                "FastQC_mqc-generalstats-fastqc-total_sequences": 600529762.0
+            },
+            "concatenated_normal_XXXXXX_R_1": {
+                "FastQC_mqc-generalstats-fastqc-percent_duplicates": 14.426654287440797,
+                "FastQC_mqc-generalstats-fastqc-total_sequences": 464581551.0
+            },
+            "concatenated_normal_XXXXXX_R_2": {
+                "FastQC_mqc-generalstats-fastqc-percent_duplicates": 14.214689357571501,
+                "FastQC_mqc-generalstats-fastqc-total_sequences": 464581551.0
+            },
+            "concatenated_tumor_XXXXXX_R_1": {
+                "FastQC_mqc-generalstats-fastqc-percent_duplicates": 15.213739762327492,
+                "FastQC_mqc-generalstats-fastqc-total_sequences": 600529762.0
+            }
         }
     }
 }