Implement pm1 vcep rules (#171)

* Another 7 VCEPs

* Another 18 VCEPs

* Another 3 VCEPs

* fix the cli test

* delete failing integration test

* Unit tests

* Some reformatting

* additional unit test

.gitignore

@@ -11,6 +11,9 @@ src/bench/tmp/*.csv
 cache/
 cache/**
 
+# Prediction file
+prediction.json
+
 # DS_Store file
 .DS_Store
 

src/auto_acmg.py

@@ -1,6 +1,6 @@
 """Implementations of the PVS1 algorithm."""
 
-from typing import Optional, Type, Union
+from typing import Dict, Optional, Type, Union
 
 from loguru import logger
 
@@ -19,7 +19,7 @@
 from src.defs.auto_acmg import AutoACMGResult, CdsInfo, GenomicStrand
 from src.defs.exceptions import AlgorithmError, AutoAcmgBaseException, ParseError
 from src.defs.genome_builds import GenomeRelease
-from src.defs.mehari import ProteinPos, TxPos
+from src.defs.mehari import CdsPos, ProteinPos, TxPos
 from src.defs.seqvar import SeqVar, SeqVarResolver
 from src.utils import SeqVarTranscriptsHelper
 from src.vcep import (
@@ -29,6 +29,34 @@
     CDH1Predictor,
     CerebralCreatineDeficiencySyndromesPredictor,
     CoagulationFactorDeficiencyPredictor,
+    CongenitalMyopathiesPredictor,
+    DICER1Predictor,
+    ENIGMAPredictor,
+    EpilepsySodiumChannelPredictor,
+    FamilialHypercholesterolemiaPredictor,
+    FBN1Predictor,
+    GlaucomaPredictor,
+    HBOPCPredictor,
+    HearingLossPredictor,
+    HHTPredictor,
+    InsightColorectalCancerPredictor,
+    LeberCongenitalAmaurosisPredictor,
+    LysosomalDiseasesPredictor,
+    MalignantHyperthermiaPredictor,
+    MitochondrialDiseasesPredictor,
+    MonogenicDiabetesPredictor,
+    MyeloidMalignancyPredictor,
+    PKUPredictor,
+    PlateletDisordersPredictor,
+    PTENPredictor,
+    PulmonaryHypertensionPredictor,
+    RASopathyPredictor,
+    RettAngelmanPredictor,
+    SCIDPredictor,
+    ThrombosisPredictor,
+    TP53Predictor,
+    VHLPredictor,
+    VonWillebrandDiseasePredictor,
 )
 
 #: Mapping of HGNC gene identifiers to predictor classes.
@@ -52,6 +80,92 @@
     "HGNC:11055": CerebralCreatineDeficiencySyndromesPredictor,  # SLC6A8
     "HGNC:3546": CoagulationFactorDeficiencyPredictor,  # F8
     "HGNC:3551": CoagulationFactorDeficiencyPredictor,  # F9
+    "HGNC:7720": CongenitalMyopathiesPredictor,  # NEB
+    "HGNC:129": CongenitalMyopathiesPredictor,  # ACTA1
+    "HGNC:2974": CongenitalMyopathiesPredictor,  # DNM2
+    "HGNC:7448": CongenitalMyopathiesPredictor,  # MTM1
+    "HGNC:10483": CongenitalMyopathiesPredictor,  # RYR1
+    "HGNC:17098": DICER1Predictor,  # DICER1
+    "HGNC:1100": ENIGMAPredictor,  # BRCA1
+    "HGNC:1101": ENIGMAPredictor,  # BRCA2
+    "HGNC:10585": EpilepsySodiumChannelPredictor,  # SCN1A
+    "HGNC:10588": EpilepsySodiumChannelPredictor,  # SCN2A
+    "HGNC:10590": EpilepsySodiumChannelPredictor,  # SCN3A
+    "HGNC:10596": EpilepsySodiumChannelPredictor,  # SCN8A
+    "HGNC:10586": EpilepsySodiumChannelPredictor,  # SCN1B
+    "HGNC:6547": FamilialHypercholesterolemiaPredictor,  # LDLR
+    "HGNC:3603": FBN1Predictor,  # FBN1
+    "HGNC:7610": GlaucomaPredictor,  # MYOC
+    "HGNC:13733": HearingLossPredictor,  # CDH23
+    "HGNC:2180": HearingLossPredictor,  # COCH
+    "HGNC:4284": HearingLossPredictor,  # GJB2
+    "HGNC:6298": HearingLossPredictor,  # KCNQ4
+    "HGNC:7605": HearingLossPredictor,  # MYO6
+    "HGNC:7606": HearingLossPredictor,  # MYO7A
+    "HGNC:8818": HearingLossPredictor,  # SLC26A4
+    "HGNC:11720": HearingLossPredictor,  # TECTA
+    "HGNC:12601": HearingLossPredictor,  # USH2A
+    "HGNC:7594": HearingLossPredictor,  # MYO15A
+    "HGNC:8515": HearingLossPredictor,  # OTOF
+    "HGNC:795": HBOPCPredictor,  # ATM
+    "HGNC:26144": HBOPCPredictor,  # PALB2
+    "HGNC:175": HHTPredictor,  # ACVRL1
+    "HGNC:3349": HHTPredictor,  # ENG
+    "HGNC:583": InsightColorectalCancerPredictor,  # APC
+    "HGNC:7127": InsightColorectalCancerPredictor,  # MLH1
+    "HGNC:7325": InsightColorectalCancerPredictor,  # MSH2
+    "HGNC:7329": InsightColorectalCancerPredictor,  # MSH6
+    "HGNC:9122": InsightColorectalCancerPredictor,  # PMS2
+    "HGNC:10294": LeberCongenitalAmaurosisPredictor,  # RPE65
+    "HGNC:4065": LysosomalDiseasesPredictor,  # GAA
+    "HGNC:10483": MalignantHyperthermiaPredictor,  # GBA
+    "HGNC:23287": MitochondrialDiseasesPredictor,  # ETHE1
+    "HGNC:8806": MitochondrialDiseasesPredictor,  # PDHA1
+    "HGNC:9179": MitochondrialDiseasesPredictor,  # POLG
+    "HGNC:16266": MitochondrialDiseasesPredictor,  # SLC19A3
+    "HGNC:11621": MonogenicDiabetesPredictor,  # HNF1A
+    "HGNC:5024": MonogenicDiabetesPredictor,  # HNF4A
+    "HGNC:4195": MonogenicDiabetesPredictor,  # GCK
+    "HGNC:10471": MyeloidMalignancyPredictor,  # RUNX1
+    "HGNC:8582": PKUPredictor,  # PAH
+    "HGNC:6138": PlateletDisordersPredictor,  # ITGA2B
+    "HGNC:6156": PlateletDisordersPredictor,  # ITGB3
+    "HGNC:9588": PTENPredictor,  # PTEN
+    "HGNC:1078": PulmonaryHypertensionPredictor,  # BMPR2
+    "HGNC:12726": VonWillebrandDiseasePredictor,  # VWF
+    "HGNC:775": ThrombosisPredictor,  # SERPINC1
+    "HGNC:11998": TP53Predictor,  # TP53
+    "HGNC:12687": VHLPredictor,  # VHL
+    "HGNC:15454": RASopathyPredictor,  # SHOC2
+    "HGNC:7989": RASopathyPredictor,  # NRAS
+    "HGNC:9829": RASopathyPredictor,  # RAF1
+    "HGNC:11187": RASopathyPredictor,  # SOS1
+    "HGNC:11188": RASopathyPredictor,  # SOS2
+    "HGNC:9644": RASopathyPredictor,  # PTPN11
+    "HGNC:6407": RASopathyPredictor,  # KRAS
+    "HGNC:6840": RASopathyPredictor,  # MAP2K1
+    "HGNC:5173": RASopathyPredictor,  # HRAS
+    "HGNC:10023": RASopathyPredictor,  # RIT1
+    "HGNC:6842": RASopathyPredictor,  # MAP2K2
+    "HGNC:1097": RASopathyPredictor,  # BRAF
+    "HGNC:7227": RASopathyPredictor,  # MRAS
+    "HGNC:6742": RASopathyPredictor,  # LZTR1
+    "HGNC:17271": RASopathyPredictor,  # RRAS2
+    "HGNC:9282": RASopathyPredictor,  # PPP1CB
+    "HGNC:11634": RettAngelmanPredictor,  # TCF4
+    "HGNC:11079": RettAngelmanPredictor,  # SLC9A6
+    "HGNC:11411": RettAngelmanPredictor,  # CDKL5
+    "HGNC:3811": RettAngelmanPredictor,  # FOXG1
+    "HGNC:6990": RettAngelmanPredictor,  # MECP2
+    "HGNC:12496": RettAngelmanPredictor,  # UBE3A
+    "HGNC:12765": SCIDPredictor,  # FOXN1
+    "HGNC:186": SCIDPredictor,  # ADA
+    "HGNC:17642": SCIDPredictor,  # DCLRE1C
+    "HGNC:6024": SCIDPredictor,  # IL7R
+    "HGNC:6193": SCIDPredictor,  # JAK3
+    "HGNC:9831": SCIDPredictor,  # RAG1
+    "HGNC:9832": SCIDPredictor,  # RAG2
+    "HGNC:6010": SCIDPredictor,  # IL2RG
 }
 
 
@@ -183,6 +297,9 @@ def parse_data(self, seqvar: SeqVar) -> AutoACMGResult:
         self.result.data.tx_pos_utr = (
             seqvar_transcript.tx_pos.ord if isinstance(seqvar_transcript.tx_pos, TxPos) else -1
         )
+        self.result.data.cds_pos = (
+            seqvar_transcript.cds_pos.ord if isinstance(seqvar_transcript.cds_pos, CdsPos) else 0
+        )
         self.result.data.prot_pos = (
             seqvar_transcript.protein_pos.ord
             if isinstance(seqvar_transcript.protein_pos, ProteinPos)

src/cli.py

@@ -54,7 +54,8 @@ def classify(
             raise InvalidGenomeBuild("Invalid genome release")
 
         auto_acmg = AutoACMG(variant, genome_release_enum)
-        auto_acmg.predict()
+        prediction = auto_acmg.predict()
+        prediction.save_to_file() if prediction else logger.error("No prediction was made.")
     except AutoAcmgBaseException as e:
         logger.error("Error occurred: {}", e)
 

src/criteria/auto_pm1.py

@@ -13,6 +13,7 @@
     AutoACMGData,
     AutoACMGPrediction,
     AutoACMGStrength,
+    GenomicStrand,
 )
 from src.defs.exceptions import AlgorithmError, AutoAcmgBaseException, InvalidAPIResposeError
 from src.defs.genome_builds import GenomeRelease
@@ -30,6 +31,44 @@ def __init__(self):
         #: comment_pm1 to store the prediction explanation.
         self.comment_pm1: str = ""
 
+    @staticmethod
+    def _get_affected_exon(var_data: AutoACMGData, seqvar: SeqVar) -> int:
+        """
+        Get the affected exon number for the variant.
+
+        Go through all exons and count them before the variant position.
+        Pay attention to the strand of the gene.
+
+        Args:
+            var_data: AutoACMGData object
+            seqvar: SeqVar object
+
+        Returns:
+            int: Affected exon number
+        """
+        exon_number = 0
+        if var_data.strand == GenomicStrand.Plus:
+            for exon in var_data.exons:
+                if exon.altStartI >= seqvar.pos:
+                    return exon_number
+                if exon.altStartI <= seqvar.pos <= exon.altEndI:
+                    exon_number += 1
+                    return exon_number
+                if exon.altEndI < seqvar.pos:
+                    exon_number += 1
+        elif var_data.strand == GenomicStrand.Minus:
+            for exon in var_data.exons[::-1]:
+                if exon.altEndI <= seqvar.pos:
+                    return exon_number
+                if exon.altStartI <= seqvar.pos <= exon.altEndI:
+                    exon_number += 1
+                    return exon_number
+                if exon.altStartI > seqvar.pos:
+                    exon_number += 1
+        else:
+            raise AlgorithmError(f"Invalid strand for {var_data.hgnc_id}: {var_data.strand}")
+        return exon_number
+
     def _count_vars(self, seqvar: SeqVar, start_pos: int, end_pos: int) -> Tuple[int, int]:
         """
         Counts pathogenic and benign variants in the specified range.

src/defs/auto_acmg.py

@@ -397,6 +397,7 @@ class AutoACMGData(AutoAcmgBaseModel):
     transcript_id: str = ""
     transcript_tags: List[str] = []
     tx_pos_utr: int = -1
+    cds_pos: int = 0
     prot_pos: int = -1
     prot_length: int = -1
     cds_info: Dict[str, CdsInfo] = {}

src/defs/core.py

@@ -30,3 +30,8 @@ class AutoAcmgBaseModel(BaseModel):
     # class Config:
     #     use_enum_values = True
     #     arbitrary_types_allowed = True
+
+    def save_to_file(self, file_path: str = "prediction.json") -> None:
+        """Save the model to a file."""
+        with open(file_path, "w") as file:
+            file.write(self.model_dump_json(indent=4))

src/vcep/__init__.py

@@ -6,3 +6,31 @@
     CerebralCreatineDeficiencySyndromesPredictor,
 )
 from src.vcep.coagulation_factor_deficiency import CoagulationFactorDeficiencyPredictor
+from src.vcep.congenital_myopathies import CongenitalMyopathiesPredictor
+from src.vcep.dicer1 import DICER1Predictor
+from src.vcep.enigma import ENIGMAPredictor
+from src.vcep.epilepsy_sodium_channel import EpilepsySodiumChannelPredictor
+from src.vcep.familial_hypercholesterolemia import FamilialHypercholesterolemiaPredictor
+from src.vcep.fbn1 import FBN1Predictor
+from src.vcep.glaucoma import GlaucomaPredictor
+from src.vcep.hbopc import HBOPCPredictor
+from src.vcep.hearing_loss import HearingLossPredictor
+from src.vcep.hhtp import HHTPredictor
+from src.vcep.insight_colorectal_cancer import InsightColorectalCancerPredictor
+from src.vcep.leber_congenital_amaurosis import LeberCongenitalAmaurosisPredictor
+from src.vcep.lysosomal_diseases import LysosomalDiseasesPredictor
+from src.vcep.malignant_hyperthermia_susceptibility import MalignantHyperthermiaPredictor
+from src.vcep.mitochondrial_diseases import MitochondrialDiseasesPredictor
+from src.vcep.monogenic_diabetes import MonogenicDiabetesPredictor
+from src.vcep.myeloid_malignancy import MyeloidMalignancyPredictor
+from src.vcep.pku import PKUPredictor
+from src.vcep.platelet_disorders import PlateletDisordersPredictor
+from src.vcep.pten import PTENPredictor
+from src.vcep.pulmonary_hypertension import PulmonaryHypertensionPredictor
+from src.vcep.rasopathy import RASopathyPredictor
+from src.vcep.rett_angelman import RettAngelmanPredictor
+from src.vcep.scid import SCIDPredictor
+from src.vcep.thrombosis import ThrombosisPredictor
+from src.vcep.tp53 import TP53Predictor
+from src.vcep.vhl import VHLPredictor
+from src.vcep.von_willebrand_disease import VonWillebrandDiseasePredictor

src/vcep/acadvl.py

@@ -10,7 +10,7 @@
 from src.defs.auto_acmg import AutoACMGCriteria, AutoACMGData, AutoACMGPrediction, AutoACMGStrength
 from src.defs.seqvar import SeqVar
 
-PM1_CLUSTER_REGIONS = [
+PM1_CLUSTER = [
     (214, 223),  # Nucleotide and substrate binding
     (249, 251),  # Nucleotide and substrate binding
     (460, 466),  # Nucleotide and substrate binding
@@ -25,7 +25,7 @@ def predict_pm1(self, seqvar: SeqVar, var_data: AutoACMGData) -> AutoACMGCriteri
         """Override predict_pm1 to include VCEP-specific logic for ACADVL."""
         logger.info("Predict PM1")
         # Check if variant falls within critical regions
-        for start, end in PM1_CLUSTER_REGIONS:
+        for start, end in PM1_CLUSTER:
             if start <= var_data.prot_pos <= end:
                 comment = (
                     f"Variant falls within a critical region for ACADVL between positions "

src/vcep/cardiomyopathy.py

@@ -1,6 +1,6 @@
 """
 Predictor for Cardiomyopathy VCEP.
-Included gene:
+Included genes:
 MYH7 (HGNC:7577),
 MYBPC3 (HGNC:7551),
 TNNI3 (HGNC:11947),

src/vcep/coagulation_factor_deficiency.py

@@ -13,13 +13,7 @@
 from loguru import logger
 
 from src.criteria.default_predictor import DefaultPredictor
-from src.defs.auto_acmg import (
-    AutoACMGCriteria,
-    AutoACMGData,
-    AutoACMGPrediction,
-    AutoACMGStrength,
-    GenomicStrand,
-)
+from src.defs.auto_acmg import AutoACMGCriteria, AutoACMGData, AutoACMGPrediction, AutoACMGStrength
 from src.defs.exceptions import AlgorithmError
 from src.defs.seqvar import SeqVar
 
@@ -39,7 +33,7 @@
         },
         "moderate": {
             "residues": [(1667, 1667), (1332, 1332)],  # Residues affecting secretion
-            "regions": [(2267, 2304)],  # FXa-binding residues, excluding Ser2283
+            "domains": [(2267, 2304)],  # FXa-binding residues, excluding Ser2283
             "excluded_residues": [(2283, 2283)],  # Excluded residue in FXa-binding region
         },
     },
@@ -60,44 +54,6 @@
 
 class CoagulationFactorDeficiencyPredictor(DefaultPredictor):
 
-    @staticmethod
-    def _get_affected_exon(var_data: AutoACMGData, seqvar: SeqVar) -> int:
-        """
-        Get the affected exon number for the variant.
-
-        Go through all exons and count them before the variant position.
-        Pay attention to the strand of the gene.
-
-        Args:
-            var_data: AutoACMGData object
-            seqvar: SeqVar object
-
-        Returns:
-            int: Affected exon number
-        """
-        exon_number = 0
-        if var_data.strand == GenomicStrand.Plus:
-            for exon in var_data.exons:
-                if exon.altStartI >= seqvar.pos:
-                    return exon_number
-                if exon.altStartI <= seqvar.pos <= exon.altEndI:
-                    exon_number += 1
-                    return exon_number
-                if exon.altEndI < seqvar.pos:
-                    exon_number += 1
-        elif var_data.strand == GenomicStrand.Minus:
-            for exon in var_data.exons[::-1]:
-                if exon.altEndI <= seqvar.pos:
-                    return exon_number
-                if exon.altStartI <= seqvar.pos <= exon.altEndI:
-                    exon_number += 1
-                    return exon_number
-                if exon.altStartI > seqvar.pos:
-                    exon_number += 1
-        else:
-            raise AlgorithmError(f"Invalid strand for {var_data.hgnc_id}: {var_data.strand}")
-        return exon_number
-
     def predict_pm1(self, seqvar: SeqVar, var_data: AutoACMGData) -> AutoACMGCriteria:
         """
         Override predict_pm1 to include VCEP-specific logic for Coagulation Factor Deficiency.
@@ -139,8 +95,8 @@ def predict_pm1(self, seqvar: SeqVar, var_data: AutoACMGData) -> AutoACMGCriteri
                     summary=comment,
                 )
 
-        # Check moderate level criteria for regions, excluding specific residues
-        for start, end in gene_cluster.get("moderate", {}).get("regions", []):
+        # Check moderate level criteria for domains, excluding specific residues
+        for start, end in gene_cluster.get("moderate", {}).get("domains", []):
             if start <= var_data.prot_pos <= end:
                 excluded_residues = gene_cluster.get("moderate", {}).get("excluded_residues", [])
                 if not any(start <= var_data.prot_pos <= end for start, end in excluded_residues):