feat: Option to support new analysis workflows

BALSAMIC/commands/config/case.py

@@ -117,13 +117,6 @@
     multiple=True,
     help="Fastq files for normal sample.",
 )
-@click.option(
-    "--umiworkflow/--no-umiworkflow",
-    default=True,
-    show_default=True,
-    is_flag=True,
-    help="Enable running UMI workflow",
-)
 @click.option("--tumor-sample-name", help="Tumor sample name")
 @click.option("--normal-sample-name", help="Normal sample name")
 @click.option(
@@ -136,6 +129,19 @@
         "will be <outdir>/genome_version"
     ),
 )
+@click.option(
+    "-w",
+    "--analysis-workflow",
+    default="balsamic",
+    show_default=True,
+    type=click.Choice(["balsamic", "balsamic-umi"]),
+    help=(
+        'Analysis workflow to run. By default: "balsamic" only '
+        "workflow will be running. If you want to run both "
+        "balsamic and UMI workflow together for panel data; "
+        'choose "balsamic-umi" option '
+    ),
+)
 @click.pass_context
 def case_config(
     context,
@@ -150,12 +156,12 @@ def case_config(
     analysis_dir,
     tumor,
     normal,
-    umiworkflow,
     tumor_sample_name,
     normal_sample_name,
     genome_version,
     balsamic_cache,
     container_version,
+    analysis_workflow,
 ):
 
     try:
@@ -190,6 +196,7 @@ def case_config(
             "analysis_dir": analysis_dir,
             "analysis_type": "paired" if normal else "single",
             "sequencing_type": "targeted" if panel_bed else "wgs",
+            "analysis_workflow": analysis_workflow,
         },
         reference=reference_dict,
         singularity=os.path.join(balsamic_cache, balsamic_version, "containers"),
@@ -207,7 +214,6 @@ def case_config(
         }
         if panel_bed
         else None,
-        umiworkflow=umiworkflow if panel_bed else False,
     ).dict(by_alias=True, exclude_none=True)
     LOG.info("Config file generated successfully")
 

BALSAMIC/commands/config/pon.py

@@ -118,6 +118,7 @@ def pon_config(
             "case_id": case_id,
             "analysis_dir": analysis_dir,
             "analysis_type": "pon",
+            "analysis_workflow": "balsamic",
             "sequencing_type": "targeted" if panel_bed else "wgs",
         },
         reference=reference_dict,

BALSAMIC/commands/config/qc.py

@@ -166,6 +166,7 @@ def qc_config(
             "case_id": case_id,
             "analysis_dir": analysis_dir,
             "analysis_type": "qc_panel",
+            "analysis_workflow": "balsamic",
             "sequencing_type": "targeted" if panel_bed else "wgs",
         },
         reference=reference_dict,
@@ -181,7 +182,6 @@ def qc_config(
         }
         if panel_bed
         else None,
-        umiworkflow=False,
     ).dict(by_alias=True, exclude_none=True)
     LOG.info("QC config file generated successfully")
 

BALSAMIC/constants/common.py

@@ -34,12 +34,12 @@
 )
 
 # Analysis related constants
+ANALYSIS_TYPES = ["paired", "single", "qc_panel", "pon"]
+ANALYSIS_WORKFLOW = ["balsamic", "balsamic-umi"]
+SEQUENCING_TYPE = ["wgs", "targeted"]
 MUTATION_CLASS = ["somatic", "germline"]
 MUTATION_TYPE = ["SNV", "SV", "CNV"]
-ANALYSIS_TYPES = ["paired", "single", "qc_panel", "pon"]
 WORKFLOW_SOLUTION = ["BALSAMIC", "Sentieon", "DRAGEN", "Sentieon_umi"]
-SEQUENCING_TYPE = ["wgs", "targeted"]
-
 
 # list of bioinfo tools for each conda env
 VALID_CONTAINER_CONDA_NAME = {

BALSAMIC/snakemake_rules/quality_control/multiqc.rule

@@ -55,7 +55,7 @@ else:
     # samtools metrics
     multiqc_input.extend(expand(bam_dir + "{sample}.samtools.{stats}.txt", sample=config["samples"], stats=['flagstats', 'idxstats', 'stats']))
 
-    if config["umiworkflow"]:
+    if config["analysis"]["analysis_workflow"]=="balsamic-umi":
         # UMI picard metrics
         multiqc_input.extend(expand(umi_qc_dir + "{sample}.umi.collect_hsmetric", sample=config["samples"]))
         multiqc_input.extend(expand(umi_qc_dir + "{sample}.umi.metrics", sample=config["samples"]))

BALSAMIC/snakemake_rules/quality_control/qc_metrics.rule

@@ -9,7 +9,7 @@ if config["analysis"]["sequencing_type"] == 'wgs':
 else:
     bcftools_counts_input.append(vep_dir + "SNV.somatic." + config["analysis"]["case_id"] + ".vardict.all.filtered.pass.stats")
 
-    if config["umiworkflow"]:
+    if config["analysis"]["analysis_workflow"]=="balsamic-umi":
         # bcftools counts
         bcftools_counts_input.append(vep_dir + "SNV.somatic." + config["analysis"]["case_id"] + ".TNscope_umi.all.filtered.pass.stats")
 

BALSAMIC/utils/models.py

@@ -14,6 +14,7 @@
     BIOINFO_TOOL_ENV,
     SEQUENCING_TYPE,
     ANALYSIS_TYPES,
+    ANALYSIS_WORKFLOW,
     WORKFLOW_SOLUTION,
     MUTATION_CLASS,
     MUTATION_TYPE,
@@ -208,7 +209,10 @@ class AnalysisModel(BaseModel):
         sequencing_type : Field(required); string literal [targeted, wgs]
             targeted : if capture kit was used to enrich specific genomic regions
             wgs : if whole genome sequencing was performed
-        analysis_dir : Field(required); existing path where to save files
+         analysis_workflow: Field(required); string literal [balsamic, balsamic-umi]
+             balsamic: execute balsamic workflow
+             balsamic-umi: execute balsamic along with UMIworkflow for panels
+         analysis_dir : Field(required); existing path where to save files
         fastq_path : Field(optional); Path where fastq files will be stored
         script : Field(optional); Path where snakemake scripts will be stored
         log : Field(optional); Path where logs will be saved
@@ -222,11 +226,13 @@ class AnalysisModel(BaseModel):
         ValueError:
             When analysis_type is set to any value other than [single, paired, pon, qc_panel]
             When sequencing_type is set to any value other than [wgs, targeted]
+            When analysis_workflow is set to any other than [balsamic, balsamic-umi]
     """
 
     case_id: str
     analysis_type: str
     sequencing_type: str
+    analysis_workflow: str
     analysis_dir: DirectoryPath
     fastq_path: Optional[DirectoryPath]
     script: Optional[DirectoryPath]
@@ -258,6 +264,15 @@ def sequencing_type_literal(cls, value) -> str:
             )
         return value
 
+    @validator("analysis_workflow", check_fields=True)
+    def analysis_workflow_literal(cls, value) -> str:
+        balsamic_analysis_workflow = ANALYSIS_WORKFLOW
+        if value not in balsamic_analysis_workflow:
+            raise ValueError(
+                f"Provided analysis workflow ({value} not supported in BALSAMIC"
+            )
+        return value
+
     @validator("analysis_dir")
     def dirpath_always_abspath(cls, value) -> str:
         return Path(value).resolve().as_posix()
@@ -426,7 +441,6 @@ class BalsamicConfigModel(BaseModel):
         singularity : Field(Path); path to singularity container of BALSAMIC
         background_variants: Field(Path(optional)); path to BACKGROUND VARIANTS for UMI
         rule_directory : Field(Path(RULE_DIRECTORY)); path where snakemake rules can be found
-        umiworkflow : Field(bool); whether UMI workflow to run in parallel
     """
 
     QC: QCModel
@@ -439,7 +453,6 @@ class BalsamicConfigModel(BaseModel):
     bioinfo_tools: dict
     bioinfo_tools_version: dict
     panel: Optional[PanelModel]
-    umiworkflow: bool
 
     @validator("reference")
     def abspath_as_str(cls, value):

BALSAMIC/workflows/balsamic.smk

@@ -213,9 +213,6 @@ if "disable_variant_caller" in config:
         if var_caller in germline_caller:
             germline_caller.remove(var_caller)
 
-LOG.info(f"The following Germline variant callers will be included in the workflow: {germline_caller}")
-LOG.info(f"The following somatic variant callers will be included in the workflow: {somatic_caller}")
-
 rules_to_include = []
 analysis_type = config['analysis']["analysis_type"]
 sequence_type = config['analysis']["sequencing_type"]
@@ -225,7 +222,16 @@ for sub,value in SNAKEMAKE_RULES.items():
     for module_name,module_rules in value.items():
       rules_to_include.extend(module_rules)
 
+if config["analysis"]["analysis_workflow"] == "balsamic":
+    rules_to_include = [rule for rule in rules_to_include if "umi" not in rule]
+    somatic_caller = [var_caller for var_caller in somatic_caller if "umi" not in var_caller]
+    somatic_caller_tmb = [var_caller for var_caller in somatic_caller_tmb if "umi" not in var_caller]
+
+
 LOG.info(f"The following rules will be included in the workflow: {rules_to_include}")
+LOG.info(f"The following Germline variant callers will be included in the workflow: {germline_caller}")
+LOG.info(f"The following somatic variant callers will be included in the workflow: {somatic_caller}")
+
 
 for r in rules_to_include:
     include: Path(RULE_DIRECTORY, r).as_posix()
@@ -277,10 +283,11 @@ if config["analysis"]["sequencing_type"] != "wgs":
         expand(vep_dir + "{vcf}.all.filtered.pass.ranked.vcf.gz", vcf=get_vcf(config, ["vardict"], [case_id]))
     )
     # UMI
-    analysis_specific_results.extend(expand(umi_qc_dir + "{sample}.umi.mean_family_depth",sample=config["samples"]))
-    if background_variant_file:
-        analysis_specific_results.extend(
-            expand(umi_qc_dir + "{case_name}.{var_caller}.AFtable.txt", case_name=case_id, var_caller=["TNscope_umi"])
+    if config["analysis"]["analysis_workflow"]=="balsamic-umi":
+        analysis_specific_results.extend(expand(umi_qc_dir + "{sample}.umi.mean_family_depth",sample=config["samples"]))
+        if background_variant_file:
+            analysis_specific_results.extend(
+                expand(umi_qc_dir + "{case_name}.{var_caller}.AFtable.txt", case_name=case_id, var_caller=["TNscope_umi"])
         )
 
 # AscatNgs

CHANGELOG.rst

@@ -1,3 +1,17 @@
+[X.X.X]
+-------
+
+Added:
+^^^^^^
+
+* New option `analysis-workflow` to balsamic config case CLI https://github.com/Clinical-Genomics/BALSAMIC/pull/932
+
+
+Changed:
+^^^^^^^^
+* UMI-workflow for panel cases to be run only with `balsamic-umi` flag https://github.com/Clinical-Genomics/BALSAMIC/issues/896
+
+
 [9.0.1]
 -------
 

tests/utils/test_models.py

@@ -261,7 +261,7 @@ def test_analysis_model():
         "analysis_type": "paired",
         "sequencing_type": "targeted",
         "analysis_dir": "tests/test_data",
-        "umiworkflow": "true",
+        "analysis_workflow": "balsamic-umi",
     }
     # THEN we can successully create a config dict
     assert AnalysisModel.parse_obj(valid_args)
@@ -272,6 +272,7 @@ def test_analysis_model():
         "analysis_type": "odd",
         "sequencing_type": "wrong",
         "analysis_dir": "tests/test_data",
+        "analysis_workflow": "umi",
     }
     # THEN should trigger ValueError
     with pytest.raises(ValueError) as excinfo: