phc-ingestion 0.8.35__tar.gz → 0.8.37__tar.gz

{phc-ingestion-0.8.35 → phc-ingestion-0.8.37}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: phc-ingestion
-Version: 0.8.35
+Version: 0.8.37
 Summary: Functions for LifeOmic PHC genomic ingestions
 License: MIT
 Author-email: LifeOmic Development <development@lifeomic.com>

{phc-ingestion-0.8.35 → phc-ingestion-0.8.37}/ingestion/nextgen/process.py

@@ -68,6 +68,7 @@ def process(
             bool(structural_path_name),
             translocations,
             hyperdiploidy_chromosomes,
+            log,
         )
         pre_filtered_somatic_vcf_path = pre_filter_somatic_vcf(
             vendor_files["somaticVcfFile"],

{phc-ingestion-0.8.35 → phc-ingestion-0.8.37}/ingestion/nextgen/util/interpretation.py

@@ -18,11 +18,12 @@ def map_vendsig(vendsig: str) -> str:
     """
     Map vendor significance for short variants
     """
-    if vendsig in ["Pathogenic"]:
+    vendsig_lower = vendsig.lower()
+    if vendsig_lower in ["pathogenic"]:
         return "VENDSIG=Pathogenic"
-    elif vendsig in ["Likely Pathogenic", "LikelyPathogenic"]:
+    elif vendsig_lower in ["likely pathogenic", "likelypathogenic"]:
         return "VENDSIG=Likely pathogenic"
-    elif vendsig in ["VUS"]:
+    elif vendsig_lower in ["vus"]:
         return "VENDSIG=Uncertain significance"
     else:
         raise RuntimeError(f"Unable to map vendor significance: {vendsig}")

phc-ingestion-0.8.37/ingestion/nextgen/util/manifest_helpers.py

@@ -0,0 +1,88 @@
+from logging import Logger
+import re
+
+
+def parse_pattern(pattern: str, line: str, name: str) -> str:
+    regex = re.compile(pattern)
+    match = regex.match(line)
+
+    if not match:
+        raise ValueError(f"Could not parse {name} from line")
+
+    return match.group(1).strip()
+
+
+def parse_indication(line: str) -> str:
+    return parse_pattern(r"^.*Reason for Referral:(.*?)(Patient|<).*$", line, "indication")
+
+
+def parse_ordering_md(line: str) -> str:
+    return parse_pattern(r"^.*Physician Name:(.*?)(Reason|<).*$", line, "ordering MD")
+
+
+def parse_patient_name(line: str) -> str:
+    return parse_pattern(r"^.*Patient Name: (.*?)(Accession|<).*$", line, "patient name")
+
+
+def parse_sample_number(line: str) -> str:
+    return parse_pattern(r"^.*Specimen #: (\d*) .*$", line, "sample number")
+
+
+def parse_body_site(line: str) -> str:
+    return parse_pattern(r"^.*Specimen:(.*?)(Age|Birthdate|<).*$", line, "body site")
+
+
+def parse_report_id(line: str) -> str:
+    return parse_pattern(r"^.*Accession #: (.*?) .*$", line, "report ID")
+
+
+def parse_report_date_single_line(line: str) -> str:
+    return parse_pattern(
+        r"^.*Diagnostic Genomics Laboratory.*(\d{2}\/\d{2}\/\d{4}).*$", line, "report date"
+    )
+
+
+def parse_report_date_multiline(patient_info_lines: list[str]) -> str:
+    in_range_trigger = False
+
+    for line in patient_info_lines:
+        if "Laboratory" in line:
+            in_range_trigger = True
+            continue
+        if in_range_trigger:
+            formatted_line = re.sub(r"<\/?T.\/?>", "", line).strip()
+            if not formatted_line:
+                continue
+            return parse_pattern(
+                r"^.*(\d{2}\/\d{2}\/\d{4}).*$", formatted_line, "report date from multiline"
+            )
+
+    raise ValueError("Could not parse report date from lines")
+
+
+def parse_report_date(patient_info_lines: list[str], log: Logger) -> str:
+    """
+    Typically, the report date is in a form like:
+    ```
+    Diagnostic Genomics Laboratory 01/01/2021
+    ```
+
+    However, sometimes the date is split across multiple lines, like:
+    ```
+    Diagnostic Genomics Laboratory
+    ...random empty lines or lines with only tags...
+    01/01/2021
+    ```
+    This function attempts to first parse the date from a single line, and if that fails,
+    it will attempt to parse it from multiple lines.
+    """
+    for line in patient_info_lines:
+        if "Laboratory" in line:
+            try:
+                report_date = parse_report_date_single_line(line)
+                return report_date
+            except ValueError:
+                log.warning("Could not parse report date from single line")
+                break
+
+    return parse_report_date_multiline(patient_info_lines)

{phc-ingestion-0.8.35 → phc-ingestion-0.8.37}/ingestion/nextgen/util/process_manifest.py

@@ -17,7 +17,7 @@ def search_and_grab(array: list, search_item: str, grab_index: int):
     return array[array.index([i for i in array if re.search(search_item, i)][0]) + grab_index]
 
 
-def extract_xml_text(xml_in_file: str):
+def extract_xml_text(xml_in_file: str) -> list[str]:
     with open(xml_in_file, "r") as f:
         xml_lines = f.readlines()
 
@@ -35,7 +35,7 @@ def extract_xml_text(xml_in_file: str):
     return patient_info_lines
 
 
-def extract_interpretation_text(xml_in_file: str):
+def extract_interpretation_text(xml_in_file: str) -> list[str]:
     with open(xml_in_file, "r") as f:
         xml_lines = f.readlines()
 
@@ -111,7 +111,7 @@ def extract_patient_data(patient_info_lines: list[str]):
     return patient_data
 
 
-def extract_test_data(patient_info_lines: list, interpretation_lines: list):
+def extract_test_data(patient_info_lines: list[str], interpretation_lines: list[str], log: Logger):
     # Initialize manifest and hard-code some values
     manifest: dict[str, Any] = {}
     manifest["testType"] = "Plasma Cell Myeloma Panel"
@@ -128,12 +128,11 @@ def extract_test_data(patient_info_lines: list, interpretation_lines: list):
     manifest["medFacilID"] = ""
     manifest["medFacilName"] = "IU Health"
 
-    for line in patient_info_lines:
-        if "reportDate" not in manifest and "Laboratory" in line:
-            report_date = manifest_helpers.parse_report_date(line)
-            manifest["reportDate"] = transform_date(report_date)
-            manifest["indexedDate"] = manifest["reportDate"]
+    report_date = manifest_helpers.parse_report_date(patient_info_lines, log)
+    manifest["reportDate"] = transform_date(report_date)
+    manifest["indexedDate"] = manifest["reportDate"]
 
+    for line in patient_info_lines:
         if "collDate" not in manifest and "Collected" in line:
             collArray = line.split(" ")
             coll_date = search_and_grab(collArray, "Collected", 1)
@@ -178,10 +177,11 @@ def process_manifest(
     include_structural: bool,
     somatic_translocations: list[str],
     hyperdiploidy_chromosomes: list[str] | None,
+    log: Logger,
 ):
     test_text = extract_xml_text(xml_in_file)
     interpretation_text = extract_interpretation_text(xml_in_file)
-    manifest = extract_test_data(test_text, interpretation_text)
+    manifest = extract_test_data(test_text, interpretation_text, log)
     manifest.update(extract_patient_data(test_text))
 
     file_prefix = f".lifeomic/nextgen/{case_id}/{case_id}"

{phc-ingestion-0.8.35 → phc-ingestion-0.8.37}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "phc-ingestion"
-version = "0.8.35"
+version = "0.8.37"
 description = "Functions for LifeOmic PHC genomic ingestions"
 authors = [
     { name = "LifeOmic Development", email = "development@lifeomic.com" },

phc-ingestion-0.8.35/ingestion/nextgen/util/manifest_helpers.py

@@ -1,42 +0,0 @@
-from logging import Logger
-import re
-
-
-def parse_pattern(pattern: str, line: str, name: str) -> str:
-    regex = re.compile(pattern)
-    match = regex.match(line)
-
-    if not match:
-        raise ValueError(f"Could not parse {name} from line")
-
-    return match.group(1).strip()
-
-
-def parse_indication(line: str) -> str:
-    return parse_pattern(r"^.*Reason for Referral:(.*?)(Patient|<).*$", line, "indication")
-
-
-def parse_ordering_md(line: str) -> str:
-    return parse_pattern(r"^.*Physician Name:(.*?)(Reason|<).*$", line, "ordering MD")
-
-
-def parse_patient_name(line: str) -> str:
-    return parse_pattern(r"^.*Patient Name: (.*?)(Accession|<).*$", line, "patient name")
-
-
-def parse_sample_number(line: str) -> str:
-    return parse_pattern(r"^.*Specimen #: (\d*) .*$", line, "sample number")
-
-
-def parse_body_site(line: str) -> str:
-    return parse_pattern(r"^.*Specimen:(.*?)(Age|Birthdate|<).*$", line, "body site")
-
-
-def parse_report_id(line: str) -> str:
-    return parse_pattern(r"^.*Accession #: (.*?) .*$", line, "report ID")
-
-
-def parse_report_date(line: str) -> str:
-    return parse_pattern(
-        r"^.*Diagnostic Genomics Laboratory.*(\d{2}\/\d{2}\/\d{4}).*$", line, "report date"
-    )