pheval 0.3.5__tar.gz → 0.3.6__tar.gz

{pheval-0.3.5 → pheval-0.3.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pheval
-Version: 0.3.5
+Version: 0.3.6
 Summary: 
 Author: Yasemin Bridges
 Author-email: y.bridges@qmul.ac.uk

{pheval-0.3.5 → pheval-0.3.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pheval"
-version = "0.3.5"
+version = "0.3.6"
 description = ""
 authors = ["Yasemin Bridges <y.bridges@qmul.ac.uk>",
   "Julius Jacobsen <j.jacobsen@qmul.ac.uk>",

{pheval-0.3.5 → pheval-0.3.6}/src/pheval/post_processing/post_processing.py

@@ -227,68 +227,7 @@ class ResultSorter:
         )
 
 
-class ScoreRanker:
-    """
-    Class for ranking scores based on a given sort order
-
-    Attributes:
-       rank (int): Represents the current rank, initialised with 0
-       current_score (float): Represents the current score, initialised with positive infinity (float("inf"))
-       count (int): Used for counting, initialised with 0
-    """
-
-    rank: int = 0
-    current_score: float = float("inf")
-    count: int = 0
-
-    def __init__(self, sort_order: SortOrder):
-        """
-        Initialise ScoreRanker
-
-        Args:
-            sort_order (SortOrder): Sorting order to be applied
-        """
-        self.sort_order = sort_order
-
-    def _check_rank_order(self, round_score: float) -> None:
-        """
-        Check if the results are correctly ordered
-
-        Args:
-            round_score (float): Score to be checked against the current score
-
-        Raises:
-            ValueError: If results are not correctly sorted.
-        """
-        if self.sort_order == SortOrder.ASCENDING and round_score < self.current_score != float(
-            "inf"
-        ):
-            raise ValueError("Results are not correctly sorted!")
-        elif self.sort_order == SortOrder.DESCENDING and round_score > self.current_score != float(
-            "inf"
-        ):
-            raise ValueError("Results are not correctly sorted!")
-
-    def rank_scores(self, round_score: float) -> int:
-        """
-        Add ranks to a result; equal scores are given the same rank, e.g., 1, 1, 3
-
-        Args:
-            round_score (float): Score to be ranked
-
-        Returns:
-            int: Rank assigned to the score
-        """
-        self._check_rank_order(round_score)
-        self.count += 1
-        if self.current_score == round_score:
-            return self.rank
-        self.current_score = round_score
-        self.rank = self.count
-        return self.rank
-
-
-def _rank_pheval_result(pheval_result: [PhEvalResult], sort_order: SortOrder) -> [PhEvalResult]:
+def _rank_pheval_result(pheval_result: [PhEvalResult], sort_order: SortOrder) -> pd.DataFrame:
     """
     Rank PhEval results post-processed from tool-specific output, managing tied scores (ex aequo)
 
@@ -297,35 +236,17 @@ def _rank_pheval_result(pheval_result: [PhEvalResult], sort_order: SortOrder) ->
         sort_order (SortOrder): Sorting order based on which ranking is performed
 
     Returns:
-        List[PhEvalResult]: Ranked PhEval results with tied scores managed
+        pd.DataFrame : Ranked PhEval results with tied scores managed
 
     Raises:
         ValueError: If an incompatible PhEval result type is encountered
     """
-    score_ranker = ScoreRanker(sort_order)
-    ranked_result = []
-    for result in pheval_result:
-        if type(result) == PhEvalGeneResult:
-            ranked_result.append(
-                RankedPhEvalGeneResult.from_gene_result(
-                    result, score_ranker.rank_scores(result.score)
-                )
-            )
-        elif type(result) == PhEvalVariantResult:
-            ranked_result.append(
-                RankedPhEvalVariantResult.from_variant_result(
-                    result, score_ranker.rank_scores(result.score)
-                )
-            )
-        elif type(result) == PhEvalDiseaseResult:
-            ranked_result.append(
-                RankedPhEvalDiseaseResult.from_disease_result(
-                    result, score_ranker.rank_scores(result.score)
-                )
-            )
-        else:
-            raise ValueError("Incompatible PhEval result type.")
-    return ranked_result
+    pheval_result_df = pd.DataFrame([data.__dict__ for data in pheval_result])
+    if sort_order == SortOrder.ASCENDING:
+        pheval_result_df["rank"] = pheval_result_df["score"].rank(method="max", ascending=True)
+    elif sort_order == SortOrder.DESCENDING:
+        pheval_result_df["rank"] = pheval_result_df["score"].rank(method="max", ascending=False)
+    return pheval_result_df
 
 
 def _return_sort_order(sort_order_str: str) -> SortOrder:
@@ -347,7 +268,7 @@ def _return_sort_order(sort_order_str: str) -> SortOrder:
         raise ValueError("Incompatible ordering method specified.")
 
 
-def _create_pheval_result(pheval_result: [PhEvalResult], sort_order_str: str) -> [PhEvalResult]:
+def _create_pheval_result(pheval_result: [PhEvalResult], sort_order_str: str) -> pd.DataFrame:
     """
     Create PhEval results with corresponding ranks based on the specified sorting order.
 
@@ -356,7 +277,7 @@ def _create_pheval_result(pheval_result: [PhEvalResult], sort_order_str: str) ->
         sort_order_str (str): String representation of the desired sorting order.
 
     Returns:
-        List[PhEvalResult]: PhEval results with ranks assigned.
+       pd.DataFrame: PhEval results with ranks assigned.
     """
     sort_order = _return_sort_order(sort_order_str)
     sorted_pheval_result = ResultSorter(pheval_result, sort_order).sort_pheval_results()
@@ -364,7 +285,7 @@ def _create_pheval_result(pheval_result: [PhEvalResult], sort_order_str: str) ->
 
 
 def _write_pheval_gene_result(
-    ranked_pheval_result: [PhEvalResult], output_dir: Path, tool_result_path: Path
+    ranked_pheval_result: pd.DataFrame, output_dir: Path, tool_result_path: Path
 ) -> None:
     """
     Write ranked PhEval gene results to a TSV file
@@ -374,8 +295,9 @@ def _write_pheval_gene_result(
         output_dir (Path): Path to the output directory
         tool_result_path (Path): Path to the tool-specific result file
     """
-    ranked_result = pd.DataFrame([data.__dict__ for data in ranked_pheval_result])
-    pheval_gene_output = ranked_result.loc[:, ["rank", "score", "gene_symbol", "gene_identifier"]]
+    pheval_gene_output = ranked_pheval_result.loc[
+        :, ["rank", "score", "gene_symbol", "gene_identifier"]
+    ]
     pheval_gene_output.to_csv(
         output_dir.joinpath(
             "pheval_gene_results/" + tool_result_path.stem + "-pheval_gene_result.tsv"
@@ -386,7 +308,7 @@ def _write_pheval_gene_result(
 
 
 def _write_pheval_variant_result(
-    ranked_pheval_result: [PhEvalResult], output_dir: Path, tool_result_path: Path
+    ranked_pheval_result: pd.DataFrame, output_dir: Path, tool_result_path: Path
 ) -> None:
     """
     Write ranked PhEval variant results to a TSV file
@@ -396,8 +318,7 @@ def _write_pheval_variant_result(
         output_dir (Path): Path to the output directory
         tool_result_path (Path): Path to the tool-specific result file
     """
-    ranked_result = pd.DataFrame([data.__dict__ for data in ranked_pheval_result])
-    pheval_variant_output = ranked_result.loc[
+    pheval_variant_output = ranked_pheval_result.loc[
         :, ["rank", "score", "chromosome", "start", "end", "ref", "alt"]
     ]
     pheval_variant_output.to_csv(
@@ -410,7 +331,7 @@ def _write_pheval_variant_result(
 
 
 def _write_pheval_disease_result(
-    ranked_pheval_result: [RankedPhEvalDiseaseResult], output_dir: Path, tool_result_path: Path
+    ranked_pheval_result: pd.DataFrame, output_dir: Path, tool_result_path: Path
 ) -> None:
     """
     Write ranked PhEval disease results to a TSV file
@@ -420,8 +341,7 @@ def _write_pheval_disease_result(
         output_dir (Path): Path to the output directory
         tool_result_path (Path): Path to the tool-specific result file
     """
-    ranked_result = pd.DataFrame([data.__dict__ for data in ranked_pheval_result])
-    pheval_disease_output = ranked_result.loc[
+    pheval_disease_output = ranked_pheval_result.loc[
         :, ["rank", "score", "disease_name", "disease_identifier"]
     ]
     pheval_disease_output.to_csv(

{pheval-0.3.5 → pheval-0.3.6}/src/pheval/prepare/create_spiked_vcf.py

@@ -328,22 +328,35 @@ class VcfSpiker:
             genotype_codes[proband_variant_data.genotype.lower()] + "\n",
         ]
 
-    def construct_vcf_records(self) -> List[str]:
+    def construct_vcf_records(self, template_vcf_name: str) -> List[str]:
         """
         Construct updated VCF records by inserting spiked variants into the correct positions within the VCF.
 
+        Args:
+            template_vcf_name (str): Name of the template VCF file.
+
         Returns:
             List[str]: Updated VCF records containing the spiked variants.
         """
         updated_vcf_records = copy(self.vcf_contents)
         for variant in self.proband_causative_variants:
-            variant = self.construct_variant_entry(variant)
-            variant_entry_position = [
+            variant_entry = self.construct_variant_entry(variant)
+            matching_indices = [
                 i
                 for i, val in enumerate(updated_vcf_records)
-                if val.split("\t")[0] == variant[0] and int(val.split("\t")[1]) < int(variant[1])
-            ][-1] + 1
-            updated_vcf_records.insert(variant_entry_position, "\t".join(variant))
+                if val.split("\t")[0] == variant_entry[0]
+                and int(val.split("\t")[1]) < int(variant_entry[1])
+            ]
+            if matching_indices:
+                variant_entry_position = matching_indices[-1] + 1
+            else:
+                info_log.warning(
+                    f"Could not find entry position for {variant.variant.chrom}-{variant.variant.pos}-"
+                    f"{variant.variant.ref}-{variant.variant.alt} in {template_vcf_name}, "
+                    "inserting at end of VCF contents."
+                )
+                variant_entry_position = len(updated_vcf_records)
+            updated_vcf_records.insert(variant_entry_position, "\t".join(variant_entry))
         return updated_vcf_records
 
     def construct_header(self, updated_vcf_records: List[str]) -> List[str]:
@@ -358,21 +371,27 @@ class VcfSpiker:
         """
         updated_vcf_file = []
         for line in updated_vcf_records:
-            text = line.replace(
-                self.vcf_header.sample_id,
-                self.proband_causative_variants[0].proband_id,
-            )
+            if line.startswith("#"):
+                text = line.replace(
+                    self.vcf_header.sample_id,
+                    self.proband_causative_variants[0].proband_id,
+                )
+            else:
+                text = line
             updated_vcf_file.append(text)
         return updated_vcf_file
 
-    def construct_vcf(self) -> List[str]:
+    def construct_vcf(self, template_vcf_name: str) -> List[str]:
         """
         Construct the entire spiked VCF file by incorporating the spiked variants into the VCF.
 
+        Args:
+            template_vcf_name (str): Name of the template VCF file.
+
         Returns:
             List[str]: The complete spiked VCF file content as a list of strings.
         """
-        return self.construct_header(self.construct_vcf_records())
+        return self.construct_header(self.construct_vcf_records(template_vcf_name))
 
 
 class VcfWriter:
@@ -454,7 +473,7 @@ def spike_vcf_contents(
             chosen_template_vcf.vcf_contents,
             phenopacket_causative_variants,
             chosen_template_vcf.vcf_header,
-        ).construct_vcf(),
+        ).construct_vcf(chosen_template_vcf.vcf_file_name),
     )
 
 

{pheval-0.3.5 → pheval-0.3.6}/src/pheval/prepare/prepare_corpus.py

@@ -39,6 +39,11 @@ def prepare_corpus(
     output_dir.joinpath("phenopackets").mkdir(exist_ok=True, parents=True)
     for phenopacket_path in all_files(phenopacket_dir):
         phenopacket_util = PhenopacketUtil(phenopacket_reader(phenopacket_path))
+        if not phenopacket_util.observed_phenotypic_features():
+            info_log.warning(
+                f"Removed {phenopacket_path.name} from the corpus due to no observed phenotypic features."
+            )
+            continue
         if variant_analysis:
             if phenopacket_util.check_incomplete_variant_record():
                 info_log.warning(