Add support to run raredisease with WES (#3232)

### Added

- Support sending exomes cases in raredisease

### Changed

- Default target bed file for both MIP and raredisease updated to "twistexomerefseq_10.2_hg19_design.bed"

### Fixed

cg/constants/constants.py

@@ -57,7 +57,7 @@ class ControlOptions(StrEnum):
     EMPTY: str = ""
 
 
-DEFAULT_CAPTURE_KIT = "twistexomerefseq_9.1_hg19_design.bed"
+DEFAULT_CAPTURE_KIT = "twistexomerefseq_10.2_hg19_design.bed"
 
 
 class CustomerId(StrEnum):

cg/meta/workflow/analysis.py

@@ -4,7 +4,7 @@
 from datetime import datetime
 from pathlib import Path
 from subprocess import CalledProcessError
-from typing import Any, Iterator
+from typing import Iterator
 
 import click
 from housekeeper.store.models import Bundle, Version
@@ -29,7 +29,7 @@
 from cg.models.cg_config import CGConfig
 from cg.models.fastq import FastqFileMeta
 from cg.services.quality_controller import QualityControllerService
-from cg.store.models import Analysis, Application, BedVersion, Case, CaseSample, Sample
+from cg.store.models import Analysis, BedVersion, Case, CaseSample, Sample
 
 LOG = logging.getLogger(__name__)
 
@@ -66,6 +66,11 @@ def process(self):
     def fastq_handler(self):
         return FastqHandler
 
+    @property
+    def is_multiple_samples_allowed(self) -> bool:
+        """Return whether the analysis supports multiple samples to be linked to the case."""
+        return True
+
     @staticmethod
     def get_help(context):
         """
@@ -370,6 +375,14 @@ def gather_file_metadata_for_sample(self, sample: Sample) -> list[FastqFileMeta]
             )
         ]
 
+    def get_validated_case(self, case_id: str) -> Case:
+        case: Case = self.status_db.get_case_by_internal_id(internal_id=case_id)
+        if not case.links:
+            raise CgError(f"No samples linked to {case_id}")
+        if nlinks := len(case.links) > 1 and not self.is_multiple_samples_allowed:
+            raise CgError(f"Only one sample per case is allowed. {nlinks} found")
+        return case
+
     def link_fastq_files_for_sample(
         self, case: Case, sample: Sample, concatenate: bool = False
     ) -> None:
@@ -692,12 +705,17 @@ def run_analysis(self, *args, **kwargs):
         raise NotImplementedError
 
     def get_data_analysis_type(self, case_id: str) -> str | None:
-        """Return data analysis type carried out."""
-        case_sample: Sample = self.status_db.get_case_samples_by_case_id(case_internal_id=case_id)[
-            0
-        ].sample
-        lims_sample: dict[str, Any] = self.lims_api.sample(case_sample.internal_id)
-        application: Application = self.status_db.get_application_by_tag(
-            tag=lims_sample.get("application")
-        )
-        return application.analysis_type if application else None
+        """
+        Return data analysis type carried out.
+        Raises:
+            ValueError: If the samples in a case have not the same analysis type.
+        """
+        case: Case = self.get_validated_case(case_id)
+        analysis_types: set[str] = {
+            link.sample.application_version.application.analysis_type for link in case.links
+        }
+        if len(analysis_types) > 1:
+            raise ValueError(
+                f"Case samples have different analysis types {', '.join(analysis_types)}"
+            )
+        return analysis_types.pop() if analysis_types else None

cg/meta/workflow/nf_analysis.py

@@ -101,11 +101,6 @@ def is_params_appended_to_nextflow_config(self) -> bool:
         """Return True if parameters should be added into the nextflow config file instead of the params file."""
         return True
 
-    @property
-    def is_multiple_samples_allowed(self) -> bool:
-        """Return whether the analysis supports multiple samples to be linked to the case."""
-        return True
-
     @property
     def is_multiqc_pattern_search_exact(self) -> bool:
         """Return True if only exact pattern search is allowed to collect metrics information from MultiQC file.
@@ -264,14 +259,10 @@ def get_sample_sheet_content_per_sample(self, case_sample: CaseSample) -> list[l
         raise NotImplementedError
 
     def get_sample_sheet_content(self, case_id: str) -> list[list[Any]]:
-        """Collect and format information required to build a sample sheet for a case.
+        """Return formatted information required to build a sample sheet for a case.
         This contains information for all samples linked to the case."""
-        case: Case = self.status_db.get_case_by_internal_id(internal_id=case_id)
-        if len(case.links) == 0:
-            raise CgError(f"No samples linked to {case_id}")
-        if nlinks := len(case.links) > 1 and not self.is_multiple_samples_allowed:
-            raise CgError(f"Only one sample per case is allowed. {nlinks} found")
-        sample_sheet_content = []
+        sample_sheet_content: list = []
+        case: Case = self.get_validated_case(case_id)
         LOG.info(f"Samples linked to case {case_id}: {len(case.links)}")
         LOG.debug("Getting sample sheet information")
         for link in case.links:

cg/meta/workflow/raredisease.py

@@ -4,15 +4,15 @@
 from pathlib import Path
 from typing import Any
 
-from cg.constants import Workflow
-from cg.constants.constants import GenomeVersion
+from cg.constants import DEFAULT_CAPTURE_KIT, Workflow
+from cg.constants.constants import AnalysisType, GenomeVersion
 from cg.constants.gene_panel import GenePanelGenomeBuild
 from cg.constants.nf_analysis import RAREDISEASE_METRIC_CONDITIONS
 from cg.constants.subject import PlinkPhenotypeStatus, PlinkSex
 from cg.meta.workflow.nf_analysis import NfAnalysisAPI
 from cg.models.cg_config import CGConfig
-from cg.models.nf_analysis import WorkflowParameters
 from cg.models.raredisease.raredisease import (
+    RarediseaseParameters,
     RarediseaseSampleSheetEntry,
     RarediseaseSampleSheetHeaders,
 )
@@ -74,11 +74,26 @@ def get_sample_sheet_content_per_sample(self, case_sample: CaseSample) -> list[l
         )
         return sample_sheet_entry.reformat_sample_content
 
-    def get_workflow_parameters(self, case_id: str) -> WorkflowParameters:
+    def get_target_bed(self, case_id: str, analysis_type: str) -> str:
+        """
+        Return the target bed file from LIMS and use default capture kit for WGS.
+        """
+        target_bed: str = self.get_target_bed_from_lims(case_id=case_id)
+        if not target_bed:
+            if analysis_type == AnalysisType.WHOLE_GENOME_SEQUENCING:
+                return DEFAULT_CAPTURE_KIT
+            raise ValueError("No capture kit was found in LIMS")
+        return target_bed
+
+    def get_workflow_parameters(self, case_id: str) -> RarediseaseParameters:
         """Return parameters."""
-        return WorkflowParameters(
+        analysis_type: AnalysisType = self.get_data_analysis_type(case_id=case_id)
+        target_bed: str = self.get_target_bed(case_id=case_id, analysis_type=analysis_type)
+        return RarediseaseParameters(
             input=self.get_sample_sheet_path(case_id=case_id),
             outdir=self.get_case_path(case_id=case_id),
+            analysis_type=analysis_type,
+            target_bed=target_bed,
         )
 
     @staticmethod

cg/models/raredisease/raredisease.py

@@ -2,7 +2,7 @@
 
 from pydantic import BaseModel
 
-from cg.models.nf_analysis import NextflowSampleSheetEntry
+from cg.models.nf_analysis import NextflowSampleSheetEntry, WorkflowParameters
 
 
 class RarediseaseQCMetrics(BaseModel):
@@ -62,3 +62,10 @@ class RarediseaseSampleSheetHeaders(StrEnum):
     @classmethod
     def list(cls) -> list[str]:
         return list(map(lambda header: header.value, cls))
+
+
+class RarediseaseParameters(WorkflowParameters):
+    """Model for Raredisease parameters."""
+
+    target_bed: str
+    analysis_type: str

tests/cli/workflow/nf_analysis/test_cli_start.py

@@ -8,6 +8,7 @@
 from cg.apps.lims import LimsAPI
 from cg.cli.workflow.base import workflow as workflow_cli
 from cg.constants import EXIT_SUCCESS, Workflow
+from cg.meta.workflow.raredisease import RarediseaseAnalysisAPI
 from cg.meta.workflow.nf_analysis import NfAnalysisAPI
 from cg.models.cg_config import CGConfig
 from tests.cli.workflow.conftest import mock_analysis_flow_cell

tests/conftest.py

@@ -2445,6 +2445,77 @@ def raredisease_deliverables_file_path(raredisease_dir, raredisease_case_id) ->
     ).with_suffix(FileExtensions.YAML)
 
 
+@pytest.fixture(scope="function")
+def raredisease_context(
+    cg_context: CGConfig,
+    helpers: StoreHelpers,
+    nf_analysis_housekeeper: HousekeeperAPI,
+    trailblazer_api: MockTB,
+    raredisease_case_id: str,
+    sample_id: str,
+    no_sample_case_id: str,
+    total_sequenced_reads_pass: int,
+    apptag_rna: str,
+    case_id_not_enough_reads: str,
+    sample_id_not_enough_reads: str,
+    total_sequenced_reads_not_pass: int,
+    mocker,
+) -> CGConfig:
+    """Context to use in CLI."""
+    cg_context.housekeeper_api_ = nf_analysis_housekeeper
+    cg_context.trailblazer_api_ = trailblazer_api
+    cg_context.meta_apis["analysis_api"] = RarediseaseAnalysisAPI(config=cg_context)
+    status_db: Store = cg_context.status_db
+
+    # Create ERROR case with NO SAMPLES
+    helpers.add_case(status_db, internal_id=no_sample_case_id, name=no_sample_case_id)
+
+    # Create textbook case with enough reads
+    case_enough_reads: Case = helpers.add_case(
+        store=status_db,
+        internal_id=raredisease_case_id,
+        name=raredisease_case_id,
+        data_analysis=Workflow.RAREDISEASE,
+    )
+
+    sample_raredisease_case_enough_reads: Sample = helpers.add_sample(
+        status_db,
+        internal_id=sample_id,
+        last_sequenced_at=datetime.now(),
+        reads=total_sequenced_reads_pass,
+        application_tag=apptag_rna,
+    )
+
+    helpers.add_relationship(
+        status_db,
+        case=case_enough_reads,
+        sample=sample_raredisease_case_enough_reads,
+    )
+
+    # Create case without enough reads
+    case_not_enough_reads: Case = helpers.add_case(
+        store=status_db,
+        internal_id=case_id_not_enough_reads,
+        name=case_id_not_enough_reads,
+        data_analysis=Workflow.RAREDISEASE,
+    )
+
+    sample_not_enough_reads: Sample = helpers.add_sample(
+        status_db,
+        internal_id=sample_id_not_enough_reads,
+        last_sequenced_at=datetime.now(),
+        reads=total_sequenced_reads_not_pass,
+        application_tag=apptag_rna,
+    )
+
+    helpers.add_relationship(status_db, case=case_not_enough_reads, sample=sample_not_enough_reads)
+
+    mocker.patch.object(RarediseaseAnalysisAPI, "get_target_bed_from_lims")
+    RarediseaseAnalysisAPI.get_target_bed_from_lims.return_value = "some_target_bed_file"
+
+    return cg_context
+
+
 @pytest.fixture(scope="function")
 def deliverable_data(raredisease_dir: Path, raredisease_case_id: str, sample_id: str) -> dict:
     return {
@@ -3882,73 +3953,6 @@ def downsample_api(
     )
 
 
-@pytest.fixture(scope="function")
-def raredisease_context(
-    cg_context: CGConfig,
-    helpers: StoreHelpers,
-    nf_analysis_housekeeper: HousekeeperAPI,
-    trailblazer_api: MockTB,
-    sample_id: str,
-    no_sample_case_id: str,
-    total_sequenced_reads_pass: int,
-    apptag_rna: str,
-    raredisease_case_id: str,
-    case_id_not_enough_reads: str,
-    sample_id_not_enough_reads: str,
-    total_sequenced_reads_not_pass: int,
-) -> CGConfig:
-    """Raredisease context to use in CLI."""
-    cg_context.housekeeper_api_ = nf_analysis_housekeeper
-    cg_context.trailblazer_api_ = trailblazer_api
-    cg_context.meta_apis["analysis_api"] = RarediseaseAnalysisAPI(config=cg_context)
-    status_db: Store = cg_context.status_db
-
-    # Create ERROR case with NO SAMPLES
-    helpers.add_case(status_db, internal_id=no_sample_case_id, name=no_sample_case_id)
-
-    # Create a textbook case with enough reads
-    case_enough_reads: Case = helpers.add_case(
-        store=status_db,
-        internal_id=raredisease_case_id,
-        name=raredisease_case_id,
-        data_analysis=Workflow.RAREDISEASE,
-    )
-
-    sample_raredisease_case_enough_reads: Sample = helpers.add_sample(
-        status_db,
-        application_tag=apptag_rna,
-        internal_id=sample_id,
-        reads=total_sequenced_reads_pass,
-        last_sequenced_at=datetime.now(),
-    )
-
-    helpers.add_relationship(
-        status_db,
-        case=case_enough_reads,
-        sample=sample_raredisease_case_enough_reads,
-    )
-
-    # Create a case without enough reads
-    case_not_enough_reads: Case = helpers.add_case(
-        store=status_db,
-        internal_id=case_id_not_enough_reads,
-        name=case_id_not_enough_reads,
-        data_analysis=Workflow.RAREDISEASE,
-    )
-
-    sample_not_enough_reads: Sample = helpers.add_sample(
-        status_db,
-        application_tag=apptag_rna,
-        internal_id=sample_id_not_enough_reads,
-        reads=total_sequenced_reads_not_pass,
-        last_sequenced_at=datetime.now(),
-    )
-
-    helpers.add_relationship(status_db, case=case_not_enough_reads, sample=sample_not_enough_reads)
-
-    return cg_context
-
-
 @pytest.fixture
 def fastq_file_meta_raw(flow_cell_name: str) -> dict:
     return {

tests/meta/workflow/test_raredisease.py

@@ -32,7 +32,7 @@ def test_get_sample_sheet_content(
     assert contains_pattern
 
 
-def test_write_params_file(raredisease_context: CGConfig, raredisease_case_id: str):
+def test_write_params_file(raredisease_context: CGConfig, raredisease_case_id: str, mocker):
     # GIVEN Raredisease analysis API and input (nextflow sample sheet path)/output (case directory) parameters
     analysis_api: RarediseaseAnalysisAPI = raredisease_context.meta_apis["analysis_api"]
 
@@ -42,6 +42,9 @@ def test_write_params_file(raredisease_context: CGConfig, raredisease_case_id: s
     # THEN care directory is created
     assert os.path.exists(analysis_api.get_case_path(case_id=raredisease_case_id))
 
+    mocker.patch.object(RarediseaseAnalysisAPI, "get_target_bed_from_lims")
+    RarediseaseAnalysisAPI.get_target_bed_from_lims.return_value = "some_target_bed_file"
+
     # WHEN writing parameters file
     analysis_api.write_params_file(case_id=raredisease_case_id)