feat: Initial implementation of Molecular bar codes handling using AGeNT

snappy_pipeline/workflows/ngs_mapping/Snakefile

@@ -118,6 +118,28 @@ rule ngs_mapping_bwa_mem2_run:
         wf.wrapper_path("bwa_mem2")
 
 
+# Run Molecular Barcodes meta-tool --------------------------------------------
+
+
+rule ngs_mapping_mbcs_run:
+    input:
+        wf.get_input_files("mbcs", "run"),
+    output:
+        **wf.get_output_files("mbcs", "run"),
+    threads: wf.get_resource("mbcs", "run", "threads")
+    resources:
+        time=wf.get_resource("mbcs", "run", "time"),
+        memory=wf.get_resource("mbcs", "run", "memory"),
+        partition=wf.get_resource("mbcs", "run", "partition"),
+        tmpdir=wf.get_resource("mbcs", "run", "tmpdir"),
+    params:
+        args=wf.substep_dispatch("mbcs", "get_args", "run"),
+    log:
+        **wf.get_log_file("mbcs", "run"),
+    wrapper:
+        wf.wrapper_path("mbcs")
+
+
 # Run STAR --------------------------------------------------------------------
 
 

snappy_pipeline/workflows/ngs_mapping/__init__.py

@@ -323,7 +323,7 @@
 EXT_NAMES = ("bam", "bai", "bam_md5", "bai_md5")
 
 #: Available read mappers for (short/Illumina) DNA-seq data
-READ_MAPPERS_DNA = ("bwa", "bwa_mem2")
+READ_MAPPERS_DNA = ("bwa", "bwa_mem2", "mbcs")
 
 #: Available read mappers for (short/Illumina) RNA-seq data
 READ_MAPPERS_RNA = ("star",)
@@ -389,6 +389,21 @@
       trim_adapters: false
       mask_duplicates: true
       split_as_secondary: true  # -M flag
+    # Configuration for mbcs (meta sub-step to deal with Molecular Barcodes & base uqality recalibration)
+    mbcs:
+      mapping_tool: REQUIRED  # Either bwa of bwa_mem2. The indices & other parameters are taken from mapper config
+      mbc_tool: agent         # Only agent currently implemented
+      agent:
+        prepare:
+          path: REQUIRED
+          lib_prep_type: REQUIRED # One of "halo" (HaloPlex), "hs" (HaloPlexHS), "xt" (SureSelect XT, XT2, XT HS), "v2" (SureSelect XT HS2) & "qxt" (SureSelect QXT)
+          extra_args: []      # Consider "-polyG 8" for NovaSeq data & "-minFractionRead 50" for 100 cycles data
+        mark_duplicates:
+          path: REQUIRED
+          consensus_mode: REQUIRED # One of "SINGLE", "HYBRID", "DUPLEX"
+          input_filter_args: []
+          consensus_filter_args: []
+          extra_args: []
     # Configuration for STAR
     star:
       path_index: REQUIRED # Required if listed in ngs_mapping.tools.rna; otherwise, can be removed.
@@ -760,6 +775,73 @@ def check_config(self):
                 )
 
 
+class MBCsStepPart(ReadMappingStepPart):
+    """Support for performing NGS alignment on MBC data"""
+
+    name = "mbcs"
+    tool_category = "dna"
+
+    LIB_PREP_TYPES = ("halo", "hs", "xt", "v2", "qxt")
+    CONSENSUS_MODES = ("SINGLE", "HYBRID", "DUPLEX")
+
+    def get_resource_usage(self, action):
+        """Get Resource Usage
+
+        :param action: Action (i.e., step) in the workflow, example: 'run'.
+        :type action: str
+
+        :return: Returns ResourceUsage for step.
+
+        :raises UnsupportedActionException: if action not in class defined list of valid actions.
+        """
+        self._validate_action(action)
+        return ResourceUsage(
+            threads=1,
+            time="24:00:00",
+            memory="4G",
+        )
+
+    def check_config(self):
+        """Check parameters in configuration.
+
+        Method checks that all parameters required to execute BWA-MEM2 are present in the
+        configuration. It further checks that the provided index has all the expected file
+        extensions. If invalid configuration, it raises InvalidConfiguration exception.
+        """
+        # Check if tool is at all included in workflow
+        if self.__class__.name not in self.config["tools"]["dna"]:
+            return  # mbcs not run, don't check configuration  # pragma: no cover
+
+        # Check mapper
+        mapper = self.config["mbcs"]["mapping_tool"]
+        assert mapper != "mbcs" and mapper in READ_MAPPERS_DNA, f'Unknown mapper "{mapper}"'
+        self.parent.sub_steps[mapper].check_config()
+
+        # Check trimmer & creak paths
+        path = self.config["mbcs"]["agent"]["prepare"]["path"]
+        if not os.path.exists(path):
+            raise InvalidConfiguration(
+                f"Expected agent's trimmer input path {path} does not exist!"
+            )
+        path = self.config["mbcs"]["agent"]["mark_duplicates"]["path"]
+        if not os.path.exists(path):
+            raise InvalidConfiguration(f"Expected agent's creak input path {path} does not exist!")
+
+        # Check mandatory options
+        option = self.config["mbcs"]["agent"]["prepare"]["lib_prep_type"]
+        if option not in self.__class__.LIB_PREP_TYPES:
+            options = '", "'.join(self.__class__.LIB_PREP_TYPES)
+            raise InvalidConfiguration(
+                f'Unkown library preparation type "{option}", valid options are "{options}"'
+            )
+        option = self.config["mbcs"]["agent"]["mark_duplicates"]["consensus_mode"]
+        if option not in self.__class__.CONSENSUS_MODES:
+            options = '", "'.join(self.__class__.CONSENSUS_MODES)
+            raise InvalidConfiguration(
+                f'Unkown consensus mode "{option}", valid options are "{options}"'
+            )
+
+
 class StarStepPart(ReadMappingStepPart):
     """Support for performing NGS alignment using STAR"""
 
@@ -1362,6 +1444,7 @@ def __init__(self, workflow, config, config_lookup_paths, config_paths, workdir)
             (
                 BwaStepPart,
                 BwaMem2StepPart,
+                MBCsStepPart,
                 ExternalStepPart,
                 LinkInStep,
                 Minimap2StepPart,

snappy_wrappers/wrappers/mbcs/environment.yaml

@@ -0,0 +1,13 @@
+channels:
+- conda-forge
+- bioconda
+dependencies:
+- cffi
+- pandas
+- numpy
+- snakemake-minimal
+- openjdk =17
+- seqtk
+- samtools
+- bwa-mem2 ==2.2.1
+- gatk4