Fix parsers

Clinical-Genomics-Lund · Jan 3, 2024 · b90fb4d · b90fb4d
1 parent 5eec581
commit b90fb4d
Show file tree

Hide file tree

Showing 5 changed files with 12 additions and 76 deletions.
diff --git a/prp/parse/phenotype/mykrobe.py b/prp/parse/phenotype/mykrobe.py
@@ -7,7 +7,7 @@
 from ...models.phenotype import PredictionSoftware as Software
 from ...models.phenotype import ResistanceGene, ResistanceVariant, VariantType
 from ...models.sample import MethodIndex
-from .utils import is_prediction_result_empty, _default_amr_phenotype
+from .utils import is_prediction_result_empty
 
 LOG = logging.getLogger(__name__)
 
@@ -45,12 +45,9 @@ def _parse_mykrobe_amr_genes(mykrobe_result) -> Tuple[ResistanceGene, ...]:
 
         gene = ResistanceGene(
             gene_symbol=element_type["variants"].split("_")[0],
-            accession=None,
             depth=depth,
-            identity=None,
             coverage=coverage,
             drugs=[element_type["drug"].lower()],
-            phenotypes=[_default_amr_phenotype()],
             element_type=ElementType.AMR,
             element_subtype=ElementAmrSubtype.AMR,
         )
@@ -99,33 +96,26 @@ def _parse_mykrobe_amr_variants(mykrobe_result) -> Tuple[ResistanceVariant, ...]
         if not element_type["susceptibility"].upper() == "R":
             continue
 
-        if element_type["variants"] is not None:
+        if element_type["variants"] is None:
             continue
 
         try:
             depth = float(element_type["genes"].split(':')[-1])
         except AttributeError:
             depth = None
 
-        var_info = element_type["variants"].split("-")[1]
+        var_info = element_type["variants"].split("-")[1].split(":")[0]
         _, ref_nt, alt_nt, position = get_mutation_type(var_info)
         var_nom = element_type["variants"].split("-")[0].split("_")[1]
         var_type, *_ = get_mutation_type(var_nom)
         variant = ResistanceVariant(
             variant_type=var_type,
-            genes=[element_type["variants"].split("_")[0]],
-            phenotypes=[_default_amr_phenotype()],
+            gene_symbol=element_type["variants"].split("_")[0],
             position=position,
             ref_nt=ref_nt,
             alt_nt=alt_nt,
             depth=depth,
-            ref_database=None,
-            ref_id=None,
-            type=None,
             change=var_nom,
-            nucleotide_change=None,
-            protein_change=None,
-            annotation=None,
             drugs=[element_type["drug"].lower()],
         )
         results.append(variant)
@@ -139,7 +129,7 @@ def parse_mykrobe_amr_pred(
     LOG.info("Parsing mykrobe prediction")
     resistance = ElementTypeResult(
         phenotypes=_get_mykrobe_amr_sr_profie(prediction),
-        genes=_parse_mykrobe_amr_genes(prediction),
+        genes=[],
         mutations=_parse_mykrobe_amr_variants(prediction),
     )
 

diff --git a/prp/parse/phenotype/resfinder.py b/prp/parse/phenotype/resfinder.py
@@ -14,7 +14,6 @@
 from ...models.phenotype import PredictionSoftware as Software
 from ...models.phenotype import ResistanceGene, ResistanceVariant, VariantType
 from ...models.sample import MethodIndex
-from .utils import _default_resistance
 
 LOG = logging.getLogger(__name__)
 
@@ -221,7 +220,7 @@ def _parse_resfinder_amr_genes(
     """Get resistance genes from resfinder result."""
     results = []
     if not "seq_regions" in resfinder_result:
-        return _default_resistance().genes
+        return [ResistanceGene()]
 
     for info in resfinder_result["seq_regions"].values():
         # Get only acquired resistance genes
@@ -327,7 +326,6 @@ def _parse_resfinder_amr_variants(
 ) -> Tuple[ResistanceVariant, ...]:
     """Get resistance genes from resfinder result."""
     results = []
-    igenes = []
     for info in resfinder_result["seq_variations"].values():
         # Get only variants from desired phenotypes
         if limit_to_phenotypes is not None:
@@ -350,9 +348,6 @@ def _parse_resfinder_amr_variants(
             var_type = VariantType.DELETION
         else:
             raise ValueError("Output has no known mutation type")
-        if not "seq_regions" in info:
-            # igenes = _default_resistance().genes
-            igenes = [""]
 
         # get gene symbol and accession nr
         gene_symbol, _, gene_accnr = info["seq_regions"][0].split(";;")
@@ -376,7 +371,6 @@ def _parse_resfinder_amr_variants(
             gene_symbol=gene_symbol,
             accession=gene_accnr,
             close_seq_name=gene_accnr,
-            genes=igenes,
             phenotypes=phenotype,
             position=info["ref_start_pos"],
             ref_nt=ref_nt,

diff --git a/prp/parse/phenotype/tbprofiler.py b/prp/parse/phenotype/tbprofiler.py
@@ -7,7 +7,6 @@
 from ...models.phenotype import PredictionSoftware as Software
 from ...models.phenotype import ResistanceVariant
 from ...models.sample import MethodIndex
-from .utils import _default_variant, _default_amr_phenotype
 
 LOG = logging.getLogger(__name__)
 
@@ -53,8 +52,8 @@ def _parse_tbprofiler_amr_variants(tbprofiler_result) -> Tuple[ResistanceVariant
 
         variant = ResistanceVariant(
             variant_type=var_type,
-            genes=[hit["gene"]],
-            phenotypes=[_default_amr_phenotype()],
+            gene_symbol=hit["gene"],
+            phenotypes=[],
             position=int(hit["genome_pos"]),
             ref_nt=hit["ref"],
             alt_nt=hit["alt"],
@@ -69,7 +68,7 @@ def _parse_tbprofiler_amr_variants(tbprofiler_result) -> Tuple[ResistanceVariant
         results.append(variant)
 
     if not results:
-        results = _default_variant().mutations
+        results = [ResistanceVariant()]
         return results
 
     return results

diff --git a/prp/parse/phenotype/utils.py b/prp/parse/phenotype/utils.py
@@ -1,55 +1,8 @@
 """Shared utility functions."""
-from ...models.phenotype import ElementTypeResult, ResistanceGene
+from ...models.phenotype import ElementTypeResult
 from ...models.phenotype import ElementType, PhenotypeInfo
 
 
-def _default_resistance() -> ElementTypeResult:
-    gene = ResistanceGene(
-        name=None,
-        virulence_category=None,
-        accession=None,
-        depth=None,
-        identity=None,
-        coverage=None,
-        ref_start_pos=None,
-        ref_end_pos=None,
-        ref_gene_length=None,
-        alignment_length=None,
-        ref_database=None,
-        phenotypes=[],
-        ref_id=None,
-        contig_id=None,
-        sequence_name=None,
-        ass_start_pos=None,
-        ass_end_pos=None,
-        strand=None,
-        element_type=None,
-        element_subtype=None,
-        target_length=None,
-        res_class=None,
-        res_subclass=None,
-        method=None,
-        close_seq_name=None,
-    )
-    genes = [
-        gene,
-    ]
-    return ElementTypeResult(phenotypes=[], genes=genes, mutations=[])
-
-
-def _default_variant() -> ElementTypeResult:
-    mutation = ResistanceGene(
-        variant_type=None,
-        genes=None,
-        phenotypes=[],
-        position=None,
-        ref_nt=None,
-        alt_nt=None,
-        depth=None,
-    )
-    mutations = [mutation]
-    return ElementTypeResult(phenotypes=[], genes=[], mutations=mutations)
-
 def _default_amr_phenotype() -> PhenotypeInfo:
     return PhenotypeInfo(
         type = ElementType.AMR,

diff --git a/prp/parse/typing.py b/prp/parse/typing.py
@@ -176,8 +176,8 @@ def parse_virulencefinder_stx_typing(path: str) -> MethodIndex | None:
                 vir_gene = parse_vir_gene(hit)
                 gene = TypingResultGeneAllele(**vir_gene.model_dump())
                 pred_result = MethodIndex(
-                    type=TypingMethod.STX, 
-                    software=Software.VIRULENCEFINDER, 
+                    type=TypingMethod.STX,
+                    software=Software.VIRULENCEFINDER,
                     result=gene
                 )
     return pred_result