Merge pull request #216 from pbashyal-nmdp/p_group_redux

P group redux

Author: mmaiers-nmdp

README.md

@@ -126,6 +126,7 @@ ard.redux('B14', 'lg')
 | Reduction Type | Description                                     |
 |----------------|-------------------------------------------------|
 | `G`            | Reduce to G Group Level                         |
+| `P`            | Reduce to P Group Level                         |
 | `lg`           | Reduce to 2 field ARD level (append `g`)        |
 | `lgx`          | Reduce to 2 field ARD level                     |
 | `W`            | Reduce/Expand to 3 field WHO nomenclature level |

api-spec.yaml

@@ -2,7 +2,7 @@ openapi: 3.0.3
 info:
   title: ARD Reduction
   description: Reduce to ARD Level
-  version: "1.0.0rc1"
+  version: "1.0.0rc2"
 servers:
   - url: 'http://localhost:8080'
 tags:

pyard/__init__.py

@@ -1,5 +1,4 @@
 # -*- coding: utf-8 -*-
-from functools import lru_cache
 
 #
 #    pyard pyARD.
@@ -24,11 +23,11 @@
 #
 from .blender import blender as dr_blender
 from .broad_splits import find_splits as find_broad_splits
+from .constants import DEFAULT_CACHE_SIZE
 from .misc import get_imgt_db_versions as db_versions
-from .misc import DEFAULT_CACHE_SIZE
 
 __author__ = """NMDP Bioinformatics"""
-__version__ = "1.0.0rc1"
+__version__ = "1.0.0rc2"
 
 
 def init(

pyard/ard.py

@@ -22,24 +22,26 @@
 #    > http://www.opensource.org/licenses/lgpl-license.php
 #
 import functools
-import sys
 import re
+import sys
 from typing import Iterable, List
 
-from . import db
-from . import data_repository as dr
 from . import broad_splits
-from .smart_sort import smart_sort_comparator
+from . import data_repository as dr
+from . import db
 from .exceptions import InvalidAlleleError, InvalidMACError, InvalidTypingError
 from .misc import (
     get_n_field_allele,
     get_2field_allele,
-    expression_chars,
-    DEFAULT_CACHE_SIZE,
+    validate_reduction_type,
+)
+from .constants import (
     HLA_regex,
     VALID_REDUCTION_TYPES,
-    validate_reduction_type,
+    expression_chars,
+    DEFAULT_CACHE_SIZE,
 )
+from .smart_sort import smart_sort_comparator
 
 default_config = {
     "reduce_serology": True,
@@ -90,9 +92,7 @@ def __init__(
         self.db_connection, _ = db.create_db_connection(data_dir, imgt_version)
 
         # Load ARS mappings
-        self.ars_mappings, p_group = dr.generate_ars_mapping(
-            self.db_connection, imgt_version
-        )
+        self.ars_mappings = dr.generate_ars_mapping(self.db_connection, imgt_version)
         # Load Alleles and XX Codes
         (
             self.valid_alleles,
@@ -101,7 +101,7 @@ def __init__(
             self.who_group,
             self.exp_alleles,
         ) = dr.generate_alleles_and_xx_codes_and_who(
-            self.db_connection, imgt_version, self.ars_mappings, p_group
+            self.db_connection, imgt_version, self.ars_mappings
         )
 
         # Generate short nulls from WHO mapping
@@ -194,6 +194,8 @@ def _redux_allele(
                 return self.ars_mappings.dup_g[allele]
             else:
                 return self.ars_mappings.g_group[allele]
+        elif redux_type == "P" and allele in self.ars_mappings.p_group:
+            return self.ars_mappings.p_group[allele]
         elif redux_type in ["lgx", "lg"]:
             if allele in self.ars_mappings.dup_lgx:
                 redux_allele = self.ars_mappings.dup_lgx[allele]

pyard/constants.py

@@ -0,0 +1,10 @@
+import re
+
+DEFAULT_CACHE_SIZE = 1_000
+
+HLA_regex = re.compile("^HLA-")
+
+VALID_REDUCTION_TYPES = ["G", "P", "lg", "lgx", "W", "exon", "U2"]
+expression_chars = ["N", "Q", "L", "S"]
+# List of P and G characters
+PandG_chars = ["P", "G"]

pyard/data_repository.py

@@ -23,7 +23,6 @@
 import copy
 import functools
 import sqlite3
-from collections import namedtuple
 
 import pandas as pd
 
@@ -37,31 +36,15 @@
     load_serology_mappings,
     load_latest_version,
 )
-from .misc import expression_chars
+from .constants import expression_chars
+from .mappings import ars_mapping_tables, ARSMapping, code_mapping_tables
 from .misc import (
     get_2field_allele,
     get_3field_allele,
     number_of_fields,
     get_1field_allele,
 )
 
-ars_mapping_tables = [
-    "dup_g",
-    "dup_lgx",
-    "g_group",
-    "lgx_group",
-    "exon_group",
-    "p_not_g",
-]
-ARSMapping = namedtuple("ARSMapping", ars_mapping_tables)
-
-code_mapping_tables = [
-    "alleles",
-    "xx_codes",
-    "who_alleles",
-    "who_group",
-]
-
 
 def expression_reduce(df):
     """
@@ -162,14 +145,22 @@ def generate_ars_mapping(db_connection: sqlite3.Connection, imgt_version):
     )
     exon_group = df_exon.set_index("A")["exon"].to_dict()
 
-    # save
-    return db.save_ars_mappings(
-        db_connection, dup_g, dup_lgx, exon_group, g_group, lgx_group, p_group, p_not_g
+    ars_mapping = ARSMapping(
+        dup_g=dup_g,
+        dup_lgx=dup_lgx,
+        g_group=g_group,
+        p_group=p_group,
+        lgx_group=lgx_group,
+        exon_group=exon_group,
+        p_not_g=p_not_g,
     )
+    db.save_ars_mappings(db_connection, ars_mapping)
+
+    return ars_mapping
 
 
 def generate_alleles_and_xx_codes_and_who(
-    db_connection: sqlite3.Connection, imgt_version, ars_mappings, p_group
+    db_connection: sqlite3.Connection, imgt_version, ars_mappings
 ):
     if db.tables_exist(db_connection, code_mapping_tables):
         return db.load_code_mappings(db_connection)
@@ -232,7 +223,7 @@ def generate_alleles_and_xx_codes_and_who(
             allele_df[["Allele", "2d"]].rename(columns={"2d": "nd"}),
             allele_df[["Allele", "3d"]].rename(columns={"3d": "nd"}),
             pd.DataFrame(ars_mappings.g_group.items(), columns=["Allele", "nd"]),
-            pd.DataFrame(p_group.items(), columns=["Allele", "nd"]),
+            pd.DataFrame(ars_mappings.p_group.items(), columns=["Allele", "nd"]),
         ],
         ignore_index=True,
     )

pyard/db.py

@@ -24,7 +24,7 @@
 import sqlite3
 from typing import Tuple, Dict, Set, List
 
-from .data_repository import ARSMapping
+from .mappings import ARSMapping
 from .misc import get_imgt_db_versions, get_default_db_directory
 
 
@@ -374,73 +374,68 @@ def load_ars_mappings(db_connection):
         db_connection, table_name="dup_lgx", columns=("allele", "lgx_group")
     )
     g_group = load_dict(db_connection, table_name="g_group", columns=("allele", "g"))
+    p_group = load_dict(db_connection, table_name="p_group", columns=("allele", "p"))
     lgx_group = load_dict(
         db_connection, table_name="lgx_group", columns=("allele", "lgx")
     )
     exon_group = load_dict(
         db_connection, table_name="exon_group", columns=("allele", "exon")
     )
     p_not_g = load_dict(db_connection, table_name="p_not_g", columns=("allele", "lgx"))
-    return (
-        ARSMapping(
-            dup_g=dup_g,
-            dup_lgx=dup_lgx,
-            g_group=g_group,
-            lgx_group=lgx_group,
-            exon_group=exon_group,
-            p_not_g=p_not_g,
-        ),
-        None,
+    return ARSMapping(
+        dup_g=dup_g,
+        dup_lgx=dup_lgx,
+        g_group=g_group,
+        p_group=p_group,
+        lgx_group=lgx_group,
+        exon_group=exon_group,
+        p_not_g=p_not_g,
     )
 
 
-def save_ars_mappings(
-    db_connection, dup_g, dup_lgx, exon_group, g_group, lgx_group, p_group, p_not_g
-):
+def save_ars_mappings(db_connection: sqlite3.Connection, ars_mapping: ARSMapping):
     save_dict(
         db_connection,
         table_name="p_not_g",
-        dictionary=p_not_g,
+        dictionary=ars_mapping.p_not_g,
         columns=("allele", "lgx"),
     )
     save_dict(
         db_connection,
         table_name="dup_g",
-        dictionary=dup_g,
+        dictionary=ars_mapping.dup_g,
         columns=("allele", "g_group"),
     )
     save_dict(
         db_connection,
         table_name="dup_lgx",
-        dictionary=dup_lgx,
+        dictionary=ars_mapping.dup_lgx,
         columns=("allele", "lgx_group"),
     )
     save_dict(
-        db_connection, table_name="g_group", dictionary=g_group, columns=("allele", "g")
+        db_connection,
+        table_name="g_group",
+        dictionary=ars_mapping.g_group,
+        columns=("allele", "g"),
+    )
+    save_dict(
+        db_connection,
+        table_name="p_group",
+        dictionary=ars_mapping.p_group,
+        columns=("allele", "p"),
     )
     save_dict(
         db_connection,
         table_name="lgx_group",
-        dictionary=lgx_group,
+        dictionary=ars_mapping.lgx_group,
         columns=("allele", "lgx"),
     )
     save_dict(
         db_connection,
         table_name="exon_group",
-        dictionary=exon_group,
+        dictionary=ars_mapping.exon_group,
         columns=("allele", "exon"),
     )
-    return (
-        ARSMapping(
-            dup_g=dup_g,
-            dup_lgx=dup_lgx,
-            g_group=g_group,
-            lgx_group=lgx_group,
-            exon_group=exon_group,
-            p_not_g=p_not_g,
-        ),
-        p_group,
-    )
 
 
 def save_code_mappings(

pyard/exceptions.py

@@ -25,6 +25,7 @@ def __str__(self) -> str:
 
 class InvalidTypingError(PyArdError):
     def __init__(self, message: str, cause=None) -> None:
+        super().__init__(message)
         self.cause = cause
 
     def __str__(self) -> str:

pyard/mappings.py

@@ -0,0 +1,20 @@
+from collections import namedtuple
+
+ars_mapping_tables = [
+    "dup_g",
+    "dup_lgx",
+    "g_group",
+    "p_group",
+    "lgx_group",
+    "exon_group",
+    "p_not_g",
+]
+code_mapping_tables = [
+    "alleles",
+    "exp_alleles",
+    "xx_codes",
+    "who_alleles",
+    "who_group",
+]
+
+ARSMapping = namedtuple("ARSMapping", ars_mapping_tables)

pyard/misc.py

@@ -1,17 +1,9 @@
 # List of expression characters
 import pathlib
-import re
 import tempfile
-from typing import List, Literal
+from typing import List
 
-HLA_regex = re.compile("^HLA-")
-
-VALID_REDUCTION_TYPES = ["G", "lg", "lgx", "W", "exon", "U2"]
-expression_chars = ["N", "Q", "L", "S"]
-# List of P and G characters
-PandG_chars = ["P", "G"]
-
-DEFAULT_CACHE_SIZE = 1_000
+from pyard.constants import VALID_REDUCTION_TYPES, expression_chars, PandG_chars
 
 
 def get_n_field_allele(allele: str, n: int, preserve_expression=False) -> str: