add strucvar dcision tree (#38)

src/seqvar_pvs1.py

@@ -20,8 +20,8 @@
 from src.defs.seqvar import SeqVar
 
 
-class SeqVarPVS1Helpers:
-    """Helper methods for PVS1 criteria for transcript."""
+class SeqVarPVS1Helper:
+    """Helper methods for PVS1 criteria for sequence variants."""
 
     @staticmethod
     def _get_pHGVS_termination(pHGVS: str) -> int:
@@ -352,7 +352,7 @@ def _choose_transcript(
         return seqvar_transcript, gene_transcript
 
 
-class SeqVarPVS1(SeqVarPVS1Helpers):
+class SeqVarPVS1(SeqVarPVS1Helper):
     """PVS1 criteria for transcript."""
 
     def __init__(self, seqvar: SeqVar):

src/strucvar_pvs1.py

@@ -1,10 +1,66 @@
 """PVS1 criteria for Structural Variants (StrucVar)."""
 
+import typer
+
 from src.defs.autopvs1 import PVS1Prediction, PVS1PredictionStrucVarPath
 from src.defs.strucvar import StrucVar, StrucVarType
 
 
-class StrucVarPVS1:
+class StrucVarPVS1Helper:
+    """Helper methods for PVS1 criteria for structural variants."""
+
+    @staticmethod
+    def _full_gene_deletion() -> bool:
+        """Check if the variant is a full gene deletion."""
+        return False
+
+    @staticmethod
+    def _deletion_disrupts_rf() -> bool:
+        """Check if the single or multiple exon deletion disrupts the reading frame."""
+        return False
+
+    @staticmethod
+    def _undergo_nmd() -> bool:
+        """Check if the deletion is expected to undergo NMD."""
+        return False
+
+    @staticmethod
+    def _in_biologically_relevant_transcript() -> bool:
+        """Check if the deletion is in a biologically relevant transcript."""
+        return False
+
+    @staticmethod
+    def _critical4protein_function() -> bool:
+        """Check if the deletion is critical for protein function."""
+        return False
+
+    @staticmethod
+    def _lof_is_frequent_in_population() -> bool:
+        """Check if loss-of-function is frequent in the population."""
+        return False
+
+    @staticmethod
+    def _lof_removes_more_then_10_percent_of_protein() -> bool:
+        """Check if the loss-of-function removes more than 10% of the protein."""
+        return False
+
+    @staticmethod
+    def _proven_in_tandem() -> bool:
+        """Check if the duplication is proven in tandem."""
+        return False
+
+    @staticmethod
+    def _presumed_in_tandem() -> bool:
+        """Check if the duplication is presumed in tandem."""
+        return False
+
+    @staticmethod
+    def _duplication_disrupts_rf() -> bool:
+        """Check if the duplication disrupts the reading frame."""
+        return False
+
+
+class StrucVarPVS1(StrucVarPVS1Helper):
     """PVS1 prediction for structural variants."""
 
     def __init__(self, variant: StrucVar):
@@ -14,21 +70,87 @@ def __init__(self, variant: StrucVar):
 
     def initialize(self):
         """Initialize the PVS1 prediction."""
-        if self.variant.sv_type == StrucVarType.DEL:
-            self.prediction_path = PVS1PredictionStrucVarPath.DEL1
-        elif self.variant.sv_type == StrucVarType.DUP:
-            self.prediction_path = PVS1PredictionStrucVarPath.DUP1
+        pass
 
     def verify_PVS1(self):
         """Verify PVS1 prediction."""
         if self.variant.sv_type == StrucVarType.DEL:
-            self.prediction = PVS1Prediction.PVS1
+            if self._full_gene_deletion():
+                self.prediction = PVS1Prediction.PVS1
+                self.prediction_path = PVS1PredictionStrucVarPath.DEL1
+            elif self._deletion_disrupts_rf() and self._undergo_nmd():
+                if self._in_biologically_relevant_transcript():
+                    self.prediction = PVS1Prediction.PVS1
+                    self.prediction_path = PVS1PredictionStrucVarPath.DEL2
+                else:
+                    self.prediction = PVS1Prediction.NotPVS1
+                    self.prediction_path = PVS1PredictionStrucVarPath.DEL3
+            elif self._deletion_disrupts_rf() and not self._undergo_nmd():
+                if self._critical4protein_function():
+                    self.prediction = PVS1Prediction.PVS1_Strong
+                    self.prediction_path = PVS1PredictionStrucVarPath.DEL4
+                else:
+                    if (
+                        self._lof_is_frequent_in_population()
+                        or not self._in_biologically_relevant_transcript()
+                    ):
+                        self.prediction = PVS1Prediction.NotPVS1
+                        self.prediction_path = PVS1PredictionStrucVarPath.DEL5
+                    else:
+                        if self._lof_removes_more_then_10_percent_of_protein():
+                            self.prediction = PVS1Prediction.PVS1_Strong
+                            self.prediction_path = PVS1PredictionStrucVarPath.DEL6
+                        else:
+                            self.prediction = PVS1Prediction.PVS1_Moderate
+                            self.prediction_path = PVS1PredictionStrucVarPath.DEL7
+            else:
+                if self._critical4protein_function():
+                    self.prediction = PVS1Prediction.PVS1_Strong
+                    self.prediction_path = PVS1PredictionStrucVarPath.DEL8
+                else:
+                    if (
+                        self._lof_is_frequent_in_population()
+                        or not self._in_biologically_relevant_transcript()
+                    ):
+                        self.prediction = PVS1Prediction.NotPVS1
+                        self.prediction_path = PVS1PredictionStrucVarPath.DEL5
+                    else:
+                        if self._lof_removes_more_then_10_percent_of_protein():
+                            self.prediction = PVS1Prediction.PVS1_Strong
+                            self.prediction_path = PVS1PredictionStrucVarPath.DEL6
+                        else:
+                            self.prediction = PVS1Prediction.PVS1_Moderate
+                            self.prediction_path = PVS1PredictionStrucVarPath.DEL7
+
         elif self.variant.sv_type == StrucVarType.DUP:
-            self.prediction = PVS1Prediction.PVS1
+            if self._proven_in_tandem():
+                if self._duplication_disrupts_rf() and self._undergo_nmd():
+                    self.prediction = PVS1Prediction.PVS1
+                    self.prediction_path = PVS1PredictionStrucVarPath.DUP1
+                else:
+                    self.prediction = PVS1Prediction.NotPVS1
+                    self.prediction_path = PVS1PredictionStrucVarPath.DUP2
+            elif self._presumed_in_tandem():
+                if self._duplication_disrupts_rf() and self._undergo_nmd():
+                    self.prediction = PVS1Prediction.PVS1_Strong
+                    self.prediction_path = PVS1PredictionStrucVarPath.DUP3
+                else:
+                    self.prediction = PVS1Prediction.NotPVS1
+                    self.prediction_path = PVS1PredictionStrucVarPath.DUP2
+
+            else:
+                self.prediction = PVS1Prediction.NotPVS1
+                self.prediction_path = PVS1PredictionStrucVarPath.DUP4
+
+        else:
+            self.prediction = PVS1Prediction.NotPVS1
+            self.prediction_path = PVS1PredictionStrucVarPath.NotSet
+            typer.secho(
+                f"Invalid structural variant. Ensure the structural variant type is either DEL or DUP.",
+                err=True,
+                fg=typer.colors.RED,
+            )
 
     def get_prediction(self) -> tuple[PVS1Prediction, PVS1PredictionStrucVarPath]:
         """Get the PVS1 prediction."""
         return self.prediction, self.prediction_path
-
-    def __repr__(self):
-        return f"{self.variant.user_repr} - {self.prediction.name}"

tests/test_seqvar_pvs1.py

@@ -11,7 +11,7 @@
 from src.defs.genome_builds import GenomeRelease
 from src.defs.mehari import GeneTranscripts, TranscriptsSeqVar
 from src.defs.seqvar import SeqVar
-from src.seqvar_pvs1 import SeqVarPVS1, SeqVarPVS1Helpers, SeqVarTranscriptsHelper
+from src.seqvar_pvs1 import SeqVarPVS1, SeqVarPVS1Helper, SeqVarTranscriptsHelper
 from tests.utils import get_json_object
 
 
@@ -70,7 +70,7 @@ def __init__(self, altStartI, altEndI, altCdsStartI=None, altCdsEndI=None, cigar
 )
 def test_get_pHGVS_termination(pHGVS, expected_termination):
     """Test the _get_pHGVS_termination method."""
-    termination = SeqVarPVS1Helpers._get_pHGVS_termination(pHGVS)
+    termination = SeqVarPVS1Helper._get_pHGVS_termination(pHGVS)
     assert termination == expected_termination, f"Failed for pHGVS: {pHGVS}"
 
 
@@ -92,7 +92,7 @@ def test_get_pHGVS_termination(pHGVS, expected_termination):
 )
 def test_undergo_nmd(exons, pHGVS, hgnc_id, expected_result):
     """Test the _undergo_nmd method."""
-    result = SeqVarPVS1Helpers()._undergo_nmd(exons, pHGVS, hgnc_id)
+    result = SeqVarPVS1Helper()._undergo_nmd(exons, pHGVS, hgnc_id)
     assert result == expected_result, f"Failed for hgnc_id: {hgnc_id}, pHGVS: {pHGVS}"
 
 
@@ -110,7 +110,7 @@ def test_undergo_nmd(exons, pHGVS, hgnc_id, expected_result):
 )
 def test_in_biologically_relevant_transcript(transcript_tags, expected_result):
     """Test the _in_biologically_relevant_transcript method."""
-    result = SeqVarPVS1Helpers._in_biologically_relevant_transcript(transcript_tags)
+    result = SeqVarPVS1Helper._in_biologically_relevant_transcript(transcript_tags)
     assert result == expected_result, f"Failed for transcript_tags: {transcript_tags}"
 
 
@@ -142,7 +142,7 @@ def test_lof_is_frequent_in_population():
 )
 def test_lof_removes_more_then_10_percent_of_protein(exons, pHGVS, expected_result):
     """Test the _lof_removes_more_then_10_percent_of_protein method."""
-    result = SeqVarPVS1Helpers._lof_removes_more_then_10_percent_of_protein(pHGVS, exons)
+    result = SeqVarPVS1Helper._lof_removes_more_then_10_percent_of_protein(pHGVS, exons)
     assert (
         result == expected_result
     ), f"Expected {expected_result} for pHGVS: {pHGVS} with exon lengths: {[exon.altEndI - exon.altStartI for exon in exons]}"