scout-browser 4.97.0__py3-none-any.whl → 4.99.0__py3-none-any.whl

scout/adapter/mongo/institute.py

@@ -60,36 +60,51 @@ class InstituteHandler(object):
     ) -> Union[dict, str]:
         """Update the information for an institute."""
 
-        add_groups = add_groups or False
+        def get_phenotype_groups() -> dict:
+            """Returns a dictionary with phenotype descriptions and abbreviations."""
+            existing_groups = (
+                institute_obj.get("phenotype_groups", PHENOTYPE_GROUPS) if add_groups else {}
+            )
+
+            if not phenotype_groups:
+                return existing_groups
+
+            group_abbreviations_list = list(group_abbreviations) if group_abbreviations else []
+
+            for i, hpo_term in enumerate(phenotype_groups):
+                hpo_obj = self.hpo_term(hpo_term)
+                if not hpo_obj:
+                    continue
+
+                existing_groups[hpo_term] = {
+                    "name": hpo_obj["description"],
+                    "abbr": group_abbreviations_list[i] if group_abbreviations_list else None,
+                }
+
+            return existing_groups
+
         institute_obj = self.institute(internal_id)
         if not institute_obj:
             raise IntegrityError("Institute {} does not exist in database".format(internal_id))
 
-        updates = {"$set": {}}
+        updates = {"$set": {}, "$unset": {}}
         updated_institute = institute_obj
 
         if sanger_recipient:
-            user_obj = self.user(sanger_recipient)
-            if not user_obj:
-                raise IntegrityError("user {} does not exist in database".format(sanger_recipient))
-
-            LOG.info(
-                "Updating sanger recipients for institute: {0} with {1}".format(
-                    internal_id, sanger_recipient
-                )
-            )
-            updates["$push"] = {"sanger_recipients": sanger_recipient}
+            old_recipients = institute_obj.get("sanger_recipients", [])
+            sanger_recipients = old_recipients + [sanger_recipient]
 
         if remove_sanger:
-            LOG.info(
-                "Removing sanger recipient {0} from institute: {1}".format(
-                    remove_sanger, internal_id
-                )
+            sanger_recipients = list(
+                set(institute_obj.get("sanger_recipients", [])) - set([remove_sanger])
             )
-            updates["$pull"] = {"sanger_recipients": remove_sanger}
 
-        # Set a number of items
-        GENERAL_SETTINGS = {
+        UPDATE_SETTINGS = {
+            "alamut_institution": alamut_institution,  # Admin setting
+            "alamut_key": alamut_key,  # Admin setting
+            "check_show_all_vars": check_show_all_vars is not None,
+            "clinvar_key": clinvar_key,  # Admin setting
+            "clinvar_submitters": clinvar_submitters,
             "cohorts": cohorts,
             "collaborators": sharing_institutes,
             "coverage_cutoff": coverage_cutoff,
@@ -98,52 +113,24 @@ class InstituteHandler(object):
             "gene_panels": gene_panels,
             "gene_panels_matching": gene_panels_matching,
             "loqusdb_id": loqusdb_ids,
+            "phenotype_groups": get_phenotype_groups(),
             "sanger_recipients": sanger_recipients,
-            "clinvar_submitters": clinvar_submitters,
-        }
-        for key, value in GENERAL_SETTINGS.items():
-            if value not in [None, ""]:
-                updates["$set"][key] = value
-
-        if phenotype_groups is not None:
-            if group_abbreviations:
-                group_abbreviations = list(group_abbreviations)
-            existing_groups = {}
-            if add_groups:
-                existing_groups = institute_obj.get("phenotype_groups", PHENOTYPE_GROUPS)
-            for i, hpo_term in enumerate(phenotype_groups):
-                hpo_obj = self.hpo_term(hpo_term)
-                if not hpo_obj:
-                    return "Term {} does not exist in database".format(hpo_term)
-                hpo_id = hpo_obj["hpo_id"]
-                description = hpo_obj["description"]
-                abbreviation = None
-                if group_abbreviations:
-                    abbreviation = group_abbreviations[i]
-                existing_groups[hpo_term] = {"name": description, "abbr": abbreviation}
-            updates["$set"]["phenotype_groups"] = existing_groups
-
-        ADMIN_SETTINGS = {
-            "alamut_key": alamut_key,
-            "alamut_institution": alamut_institution,
-            "clinvar_key": clinvar_key,
-            "show_all_cases_status": show_all_cases_status,
-            "soft_filters": soft_filters,
+            "show_all_cases_status": show_all_cases_status,  # Admin setting
+            "soft_filters": soft_filters,  # Admin setting
         }
-        for key, value in ADMIN_SETTINGS.items():
-            if value not in [None, "", []]:
+        for key, value in UPDATE_SETTINGS.items():
+            if bool(value) is True:
                 updates["$set"][key] = value
+            else:
+                updates["$unset"][key] = ""  # Remove the key from the institute document
 
-        updates["$set"]["check_show_all_vars"] = check_show_all_vars is not None
-
-        if updates["$set"].keys() or updates.get("$push") or updates.get("$pull"):
+        if any(updates.get(op) for op in ["$set", "$unset"]):
             updates["$set"]["updated_at"] = datetime.now()
             updated_institute = self.institute_collection.find_one_and_update(
                 {"_id": internal_id},
                 updates,
                 return_document=pymongo.ReturnDocument.AFTER,
             )
-
             LOG.info("Institute updated")
 
         return updated_institute

scout/adapter/mongo/variant.py

@@ -2,7 +2,7 @@
 # stdlib modules
 import logging
 import re
-from typing import Any
+from typing import Any, Dict, Iterable
 
 # Third party modules
 import pymongo
@@ -689,25 +689,30 @@ class VariantHandler(VariantLoader):
         result = self.variant_collection.delete_many(query)
         LOG.info("{0} variants deleted".format(result.deleted_count))
 
-    def overlapping(self, variant_obj, limit=30):
+    def hgnc_overlapping(self, variant_obj: dict, limit: int = None) -> Iterable[Dict]:
         """Return overlapping variants.
 
-        Look at the genes that a variant overlaps to.
-        Then return all variants that overlap these genes.
+        Look at the genes that a variant overlaps, then return all variants that overlap these genes.
 
-        If variant_obj is sv it will return the overlapping snvs and oposite
-        There is a problem when SVs are huge since there are to many overlapping variants.
+        The operation is slightly different depending on the category of the variants that we want to collect.
+        If variant_obj is an SV it will return the hgnc_id matching SVs, SNVs, and MEIs but
+        for SNVs we will only return the SVs and MEIs since the genmod compounds are way better.
 
-        Args:
-            variant_obj(dict)
+        Do not return the present variant as matching.
 
-        Returns:
-            variants(iterable(dict))
+        limit: A maximum count of returned variants is introduced: mainly this is a problem when SVs are huge since there can be many genes and overlapping variants.
+               We sort to offer the LIMIT most severe overlapping variants.
         """
-        # This is the category of the variants that we want to collect
-        category = "snv" if variant_obj["category"] == "sv" else "sv"
+        category = (
+            {"$in": ["sv", "mei"]}
+            if variant_obj["category"] == "snv"
+            else {"$in": ["sv", "snv", "mei", "cancer", "cancer_sv"]}
+        )
+
         variant_type = variant_obj.get("variant_type", "clinical")
         hgnc_ids = variant_obj["hgnc_ids"]
+        if not limit:
+            limit = 30 if variant_obj["category"] == "snv" else 45
 
         query = {
             "$and": [
@@ -715,13 +720,12 @@ class VariantHandler(VariantLoader):
                 {"category": category},
                 {"variant_type": variant_type},
                 {"hgnc_ids": {"$in": hgnc_ids}},
+                {"variant_id": {"$ne": variant_obj["variant_id"]}},
             ]
         }
         sort_key = [("rank_score", pymongo.DESCENDING)]
-        # We collect the 30 most severe overlapping variants
-        variants = self.variant_collection.find(query).sort(sort_key).limit(limit)
 
-        return variants
+        return self.variant_collection.find(query).sort(sort_key).limit(limit)
 
     def evaluated_variant_ids_from_events(self, case_id, institute_id):
         """Returns variant ids for variants that have been evaluated

scout/adapter/mongo/variant_loader.py

@@ -621,16 +621,16 @@ class VariantLoader(object):
 
     def load_variants(
         self,
-        case_obj,
-        variant_type="clinical",
-        category="snv",
-        rank_threshold=None,
-        chrom=None,
-        start=None,
-        end=None,
-        gene_obj=None,
-        custom_images=None,
-        build="37",
+        case_obj: dict,
+        variant_type: str = "clinical",
+        category: str = "snv",
+        rank_threshold: float = None,
+        chrom: str = None,
+        start: int = None,
+        end: int = None,
+        gene_obj: dict = None,
+        custom_images: list = None,
+        build: str = "37",
     ):
         """Load variants for a case into scout.
 
@@ -675,7 +675,7 @@ class VariantLoader(object):
             )
 
         gene_to_panels = self.gene_to_panels(case_obj)
-        genes = [gene_obj for gene_obj in self.all_genes(build=build)]
+        genes = list(self.all_genes(build=build))
         hgncid_to_gene = self.hgncid_to_gene(genes=genes, build=build)
         genomic_intervals = self.get_coding_intervals(genes=genes, build=build)
 

scout/build/individual.py

@@ -6,9 +6,11 @@ from scout.exceptions import PedigreeError
 
 log = logging.getLogger(__name__)
 BUILD_INDIVIDUAL_FILES = [
+    "assembly_alignment_path",
     "bam_file",
     "d4_file",
     "mt_bam",
+    "paraphase_alignment_path",
     "rhocall_bed",
     "rhocall_wig",
     "rna_alignment_path",

scout/build/variant/variant.py

@@ -1,5 +1,6 @@
 # -*- coding: utf-8 -*-
 import logging
+from typing import List
 
 from scout.utils.convert import call_safe
 from scout.utils.dict_utils import remove_nonetype
@@ -115,6 +116,7 @@ def build_variant(
         revel_score = float, REVEL rankscore
         revel = float, REVEL score
         clnsig = list, # list of <clinsig>
+
         spidex = float,
 
         missing_data = bool, # default False
@@ -247,6 +249,7 @@ def build_variant(
     add_hgnc_symbols(variant_obj, variant_obj["hgnc_ids"], hgncid_to_gene)
     link_gene_panels(variant_obj, gene_to_panels)
     add_clnsig_objects(variant_obj, variant.get("clnsig", []))
+    add_clnsig_onc_objects(variant_obj, variant.get("clnsig_onc"))
 
     add_callers(variant_obj, variant.get("callers", {}))
 
@@ -337,6 +340,11 @@ def add_clnsig_objects(variant_obj, clnsig_list):
         variant_obj["clnsig"] = clnsig_objects
 
 
+def add_clnsig_onc_objects(variant_obj: dict, onc_clnsig: List[dict]):
+    if onc_clnsig:
+        variant_obj["clnsig_onc"] = onc_clnsig
+
+
 def add_callers(variant_obj, call_info):
     """Add call_info to variant_obj
     Args: variant_obj (Dict)

scout/commands/download/ensembl.py

@@ -8,9 +8,20 @@ import click
 
 from scout.utils.ensembl_biomart_clients import EnsemblBiomartHandler
 
+CHROM_SEPARATOR = "[success]"
+NR_EXPECTED_CHROMS = 24
+
 LOG = logging.getLogger(__name__)
 
 
+def integrity_check(nr_chromosomes_in_file: int):
+    if nr_chromosomes_in_file < NR_EXPECTED_CHROMS:
+        raise BufferError(
+            "Ensembl resource does not seem to be complete. Please retry downloading genes/transcripts."
+        )
+    LOG.info("Integrity check OK.")
+
+
 def print_ensembl(
     out_dir: pathlib.Path, resource_type: List[str], genome_build: Optional[str] = None
 ):
@@ -31,14 +42,19 @@ def print_ensembl(
 
         file_name: str = f"ensembl_{resource_type}_{build}.txt"
         file_path = out_dir / file_name
+        nr_chroms_in_file = 0
 
         LOG.info("Print ensembl info %s to %s", build, file_path)
 
         with file_path.open("w", encoding="utf-8") as outfile:
             for line in ensembl_client.stream_resource(interval_type=resource_type):
-                outfile.write(line + "\n")
+                if line.strip() == CHROM_SEPARATOR:
+                    nr_chroms_in_file += 1
+                else:
+                    outfile.write(line + "\n")
 
-        LOG.info(f"{file_name} file saved to disk")
+        LOG.info(f"{file_name} file saved to disk.")
+        integrity_check(nr_chroms_in_file)
 
 
 @click.command("ensembl", help="Download files with ensembl info")

scout/commands/update/individual.py

@@ -7,6 +7,7 @@ import click
 from scout.server.extensions import store
 
 UPDATE_DICT = {
+    "assembly_alignment_path": "path",
     "bam_file": "path",
     "bionano_access.sample": "str",
     "bionano_access.project": "str",
@@ -16,6 +17,7 @@ UPDATE_DICT = {
     "chromograph_images.upd_regions": "str",
     "chromograph_images.upd_sites": "str",
     "mt_bam": "path",
+    "paraphase_alignment_path": "path",
     "reviewer.alignment": "path",
     "reviewer.alignment_index": "path",
     "reviewer.vcf": "path",

scout/commands/update/panelapp.py

@@ -5,6 +5,7 @@ import logging
 import click
 from flask.cli import current_app, with_appcontext
 
+from scout.constants.panels import PANELAPPGREEN_DISPLAY_NAME, PANELAPPGREEN_NAME
 from scout.load.panelapp import load_panelapp_green_panel
 from scout.server.extensions import store
 
@@ -31,8 +32,15 @@ LOG = logging.getLogger(__name__)
     is_flag=True,
     help="Force update even if updated panel contains less genes",
 )
+@click.option("--panel-id", help="Panel ID", default=PANELAPPGREEN_NAME, show_default=True)
+@click.option(
+    "--panel-display-name",
+    help="Panel display name",
+    default=PANELAPPGREEN_DISPLAY_NAME,
+    show_default=True,
+)
 @with_appcontext
-def panelapp_green(institute, force, signed_off):
+def panelapp_green(institute, force, signed_off, panel_id, panel_display_name):
     """
     Update the automatically generated PanelApp Green Genes panel in the database.
     """
@@ -47,7 +55,12 @@ def panelapp_green(institute, force, signed_off):
 
     try:
         load_panelapp_green_panel(
-            adapter=store, institute=institute, force=force, signed_off=signed_off
+            adapter=store,
+            institute=institute,
+            force=force,
+            signed_off=signed_off,
+            panel_id=panel_id,
+            panel_display_name=panel_display_name,
         )
     except Exception as err:
         LOG.error(err)

scout/constants/__init__.py

@@ -45,7 +45,11 @@ from .disease_parsing import (
     MIMNR_PATTERN,
     OMIM_STATUS_MAP,
 )
-from .file_types import ORDERED_FILE_TYPE_MAP, ORDERED_OMICS_FILE_TYPE_MAP
+from .file_types import (
+    DNA_SAMPLE_VARIANT_CATEGORIES,
+    ORDERED_FILE_TYPE_MAP,
+    ORDERED_OMICS_FILE_TYPE_MAP,
+)
 from .filters import (
     CLINICAL_FILTER_BASE,
     CLINICAL_FILTER_BASE_CANCER,
@@ -64,12 +68,7 @@ from .gene_tags import (
     PANEL_GENE_INFO_TRANSCRIPTS,
     UPDATE_GENES_RESOURCES,
 )
-from .igv_tracks import (
-    CASE_SPECIFIC_TRACKS,
-    HUMAN_REFERENCE,
-    IGV_TRACKS,
-    USER_DEFAULT_TRACKS,
-)
+from .igv_tracks import CASE_SPECIFIC_TRACKS, HUMAN_REFERENCE, IGV_TRACKS, USER_DEFAULT_TRACKS
 from .indexes import ID_PROJECTION, INDEXES
 from .panels import PANELAPP_CONFIDENCE_EXCLUDE
 from .phenotype import (

scout/constants/clnsig.py

@@ -65,3 +65,5 @@ TRUSTED_REVSTAT_LEVEL = [
     "guideline",
     "practice_guideline",
 ]
+
+ONC_CLNSIG = ["Oncogenic", "Likely oncogenic", "Uncertain significance", "Likely benign", "Benign"]

scout/constants/file_types.py

@@ -23,6 +23,18 @@ ORDERED_FILE_TYPE_MAP = OrderedDict(
     ]
 )
 
+DNA_SAMPLE_VARIANT_CATEGORIES = [
+    "snv",
+    "sv",
+    "mei",
+    "str",
+    "vcf_snv_mt",
+    "vcf_snv_research_mt",
+    "vcf_snv_research",
+    "vcf_sv_research_mt",
+    "vcf_sv_research",
+    "vcf_mei_research",
+]
 
 ORDERED_OMICS_FILE_TYPE_MAP = OrderedDict(
     [

scout/constants/igv_tracks.py

@@ -11,7 +11,7 @@ HG38REF_INDEX_URL = "https://s3.amazonaws.com/igv.broadinstitute.org/genomes/seq
 HG38ALIAS_URL = "https://igv.org/genomes/data/hg38/hg38_alias.tab"
 HG38CYTOBAND_URL = "https://igv-genepattern-org.s3.amazonaws.com/genomes/hg38/cytoBandIdeo.txt.gz"
 
-HG38GENES_URL = "https://s3.amazonaws.com/igv.org.genomes/hg38/refGene.txt.gz"
+HG38GENES_URL = "https://hgdownload.soe.ucsc.edu/goldenPath/hg38/database/ncbiRefSeq.txt.gz"
 HG38GENES_FORMAT = "refgene"
 
 HG19GENES_URL = "https://s3.amazonaws.com/igv.org.genomes/hg19/ncbiRefSeq.sorted.txt.gz"
@@ -124,11 +124,13 @@ HUMAN_GENES_38 = {
 }
 
 CASE_SPECIFIC_TRACKS = {
-    "rhocall_bed": "Rhocall Zygosity",
-    "rhocall_wig": "Rhocall Regions",
-    "tiddit_coverage_wig": "TIDDIT Coverage",
-    "upd_regions_bed": "UPD regions",
-    "upd_sites_bed": "UPD sites",
+    "paraphase_alignments": "Paraphase Alignment",
+    "assembly_alignments": "de novo Assembly Alignment",
+    "rhocall_beds": "Rhocall Zygosity",
+    "rhocall_wigs": "Rhocall Regions",
+    "tiddit_coverage_wigs": "TIDDIT Coverage",
+    "upd_regions_beds": "UPD regions",
+    "upd_sites_beds": "UPD sites",
 }
 
 HUMAN_REFERENCE = {"37": HUMAN_REFERENCE_37, "38": HUMAN_REFERENCE_38}

scout/constants/panels.py

@@ -25,3 +25,6 @@ PRESELECTED_PANELAPP_PANEL_TYPE_SLUGS = [
     "gms-signed-off",
     "rare-disease-100k",
 ]
+
+PANELAPPGREEN_NAME = "PANELAPP-GREEN"
+PANELAPPGREEN_DISPLAY_NAME = "PanelApp Green Genes"

scout/constants/variant_tags.py

@@ -244,7 +244,7 @@ CANCER_TIER_OPTIONS = {
     "4": {
         "label": "Tier IV",
         "description": "Observed at high frequency in the population. No published evidence.",
-        "label_class": "default",
+        "label_class": "success",
     },
 }
 
@@ -334,7 +334,7 @@ MANUAL_RANK_OPTIONS = OrderedDict(
                 "label": "VUS",
                 "name": "Unknown Significance",
                 "description": "Variant of unknown significance",
-                "label_class": "default",
+                "label_class": "primary",
             },
         ),
         (
@@ -361,7 +361,7 @@ MANUAL_RANK_OPTIONS = OrderedDict(
                 "label": "RF",
                 "name": "Risk Factor",
                 "description": "Established risk allele - strong evidence for a small risk increase",
-                "label_class": "default",
+                "label_class": "dark",
             },
         ),
         (
@@ -370,7 +370,7 @@ MANUAL_RANK_OPTIONS = OrderedDict(
                 "label": "LRF",
                 "name": "Likely Risk Factor",
                 "description": "Likely risk allele - some evidence for a small risk increase",
-                "label_class": "default",
+                "label_class": "dark",
             },
         ),
         (
@@ -379,7 +379,7 @@ MANUAL_RANK_OPTIONS = OrderedDict(
                 "label": "URF",
                 "name": "Uncertain Risk Factor",
                 "description": "Uncertain risk allele - uncertain evidence for a small risk increase",
-                "label_class": "default",
+                "label_class": "dark",
             },
         ),
         (
@@ -388,7 +388,7 @@ MANUAL_RANK_OPTIONS = OrderedDict(
                 "label": "O",
                 "name": "Other",
                 "description": "Other, phenotype not related to disease",
-                "label_class": "default",
+                "label_class": "dark",
             },
         ),
     ]

scout/demo/643594.config.yaml

@@ -88,6 +88,7 @@ samples:
     splice_junctions_bed: scout/demo/ACC5963A1_lanes_1234_star_sorted_sj_filtered_sorted.bed.gz
     d4_file: scout/demo/ADM1059A3.d4
 
+
 custom_images:
   str_variants_images:
     - title: A png image

scout/load/panelapp.py

@@ -11,7 +11,6 @@ from scout.parse.panelapp import parse_panelapp_panel
 from scout.server.extensions import panelapp
 
 LOG = logging.getLogger(__name__)
-PANEL_NAME = "PANELAPP-GREEN"
 
 
 def load_panelapp_panel(
@@ -72,7 +71,14 @@ def get_panelapp_genes(
     return genes
 
 
-def load_panelapp_green_panel(adapter: MongoAdapter, institute: str, force: bool, signed_off: bool):
+def load_panelapp_green_panel(
+    adapter: MongoAdapter,
+    institute: str,
+    force: bool,
+    signed_off: bool,
+    panel_id: str,
+    panel_display_name: str,
+):
     """Load/Update the panel containing all Panelapp Green genes."""
 
     def parse_types_filter(types_filter: str, available_types: List[str]) -> List[str]:
@@ -85,10 +91,10 @@ def load_panelapp_green_panel(adapter: MongoAdapter, institute: str, force: bool
         return [available_types[i] for i in index_list]
 
     # check and set panel version
-    old_panel = adapter.gene_panel(panel_id=PANEL_NAME)
+    old_panel = adapter.gene_panel(panel_id=panel_id)
     green_panel = {
-        "panel_name": PANEL_NAME,
-        "display_name": "PanelApp Green Genes",
+        "panel_name": panel_id,
+        "display_name": panel_display_name,
         "institute": institute,
         "version": float(math.floor(old_panel["version"]) + 1) if old_panel else 1.0,
         "date": datetime.now(),

scout/models/case/case_loading_models.py

@@ -25,9 +25,11 @@ LOG = logging.getLogger(__name__)
 REPID = "{REPID}"
 
 SAMPLES_FILE_PATH_CHECKS = [
+    "assembly_alignment_path",
     "bam_file",
     "d4_file",
     "mitodel_file",
+    "paraphase_alignment_path",
     "rhocall_bed",
     "rhocall_wig",
     "rna_alignment_path",
@@ -201,6 +203,7 @@ class REViewer(BaseModel):
 
 class SampleLoader(BaseModel):
     alignment_path: Optional[str] = None
+    assembly_alignment_path: Optional[str] = None
     analysis_type: Literal[ANALYSIS_TYPES] = None
     bam_file: Optional[str] = ""
     bam_path: Optional[str] = None
@@ -221,6 +224,7 @@ class SampleLoader(BaseModel):
     mother: Optional[str] = None
     msi: Optional[str] = None
     mt_bam: Optional[str] = None
+    paraphase_alignment_path: Optional[str] = None
     phenotype: Literal["affected", "unaffected", "unknown"]
     predicted_ancestry: Optional[str] = None
     reviewer: Optional[REViewer] = REViewer()

scout/parse/variant/clnsig.py

@@ -3,9 +3,47 @@ from typing import Dict, List, Optional, Union
 
 import cyvcf2
 
+from scout.constants.clnsig import ONC_CLNSIG
+
 LOG = logging.getLogger(__name__)
 
 
+def split_groups(value: str) -> List[str]:
+    """Removes leading underscore from a string and splits it into a list of items."""
+    return [
+        item.lstrip("_").replace(" ", "_")
+        for group in value.replace("&", ",").split(",")
+        for item in group.split("/")
+    ]
+
+
+def parse_clnsig_onc(variant: cyvcf2.Variant) -> List[dict]:
+    """Collect somatic oncogenicity ClinVar classifications for a variant, if available."""
+    if not variant.INFO.get("ONC"):
+        return []
+    acc = int(variant.INFO.get("CLNVID", 0))
+    onc_sig_groups = split_groups(value=variant.INFO.get("ONC", "").lower())
+    onc_revstat = ",".join(split_groups(value=variant.INFO.get("ONCREVSTAT", "").lower()))
+    onc_dn_groups = split_groups(variant.INFO.get("ONCDN", ""))
+
+    onc_clnsig_accessions = []
+
+    for i, onc_sig in enumerate(onc_sig_groups):
+        if (
+            onc_sig.capitalize() not in ONC_CLNSIG
+        ):  # This is excluding entries with ONC=no_classification_for_the_single_variant
+            continue
+        onc_clnsig_accessions.append(
+            {
+                "accession": acc,
+                "value": onc_sig,
+                "revstat": onc_revstat,
+                "dn": onc_dn_groups[i].replace("|", ","),
+            }
+        )
+    return onc_clnsig_accessions
+
+
 def parse_clnsig_low_penetrance(sig_groups: List[str]) -> List[str]:
     """If 'low_penetrance' is among the clnsig terms of an array, the term gets appended to the term immediately before in the array."""
     result = []