Allow to Refresh MAC data

  - adds `refresh_mac` flag to ARD() to allow updating of MAC data

Author: pbashyal-nmdp

README.rst

@@ -54,17 +54,21 @@ Example
 
 .. code-block:: python3
 
-    from pyard import ARD
+    import pyard
 
-    # Initialize ARD object
-    ard = ARD('3290')
+    # Initialize ARD object with a version of IMGT HLA database
+    ard = pyard.ARD(3290)
 
     # You can specify a data directory for temp files
-    # ard = ARD('3290', data_dir='/tmp/py-ard')
+    # ard = pyard.ARD('3290', data_dir='/tmp/py-ard')
 
-    # Initialize with latest DB
-    ard = ARD()
+    # Initialize with latest IMGT HLA database
+    ard = pyard.ARD()
 
+    # You can choose to refresh the MAC code for previously used versions
+    # ard =  pyard.ARD(3290, refresh_mac=True)
+
+    # Allele to reduce
     allele = "A*01:01:01"
 
     ard.redux(allele, 'G')

pyard/data_repository.py

@@ -97,7 +97,7 @@ def generate_ars_mapping(db_connection: sqlite3.Connection, imgt_version):
     return dup_g, g_group, lg_group, lgx_group
 
 
-def generate_mac_codes(db_connection: sqlite3.Connection):
+def generate_mac_codes(db_connection: sqlite3.Connection, refresh_mac: bool):
     """
     MAC files come in 2 different versions:
 
@@ -144,14 +144,16 @@ def generate_mac_codes(db_connection: sqlite3.Connection):
     :return:
     """
     mac_table_name = 'mac_codes'
-    if not db.table_exists(db_connection, mac_table_name):
+    if refresh_mac or not db.table_exists(db_connection, mac_table_name):
         # Load the MAC file to a DataFrame
         mac_url = 'https://hml.nmdp.org/mac/files/numer.v3.zip'
-        df_mac = pd.read_csv(mac_url, sep='\t', compression='zip', skiprows=3, names=['Code', 'Alleles'])
+        df_mac = pd.read_csv(mac_url, sep='\t', compression='zip',
+                             skiprows=3, names=['Code', 'Alleles'])
         # Create a dict from code to alleles
         mac = df_mac.set_index("Code")["Alleles"].to_dict()
         # Save the mac dict to db
-        db.save_dict(db_connection, table_name=mac_table_name, dictionary=mac, columns=('code', 'alleles'))
+        db.save_dict(db_connection, table_name=mac_table_name,
+                     dictionary=mac, columns=('code', 'alleles'))
 
 
 def generate_alleles_and_xx_codes(db_connection: sqlite3.Connection, imgt_version):

pyard/db.py

@@ -110,12 +110,16 @@ def save_dict(connection: sqlite3.Connection, table_name: str,
     :return: success status
     """
     cursor = connection.cursor()
+
+    # Drop the table first
+    drop_table_sql = f"DROP TABLE IF EXISTS {table_name}"
+    cursor.execute(drop_table_sql)
+
+    # Create table
     create_table_sql = f"""CREATE TABLE {table_name} (
                             {columns[0]} TEXT PRIMARY KEY,
                             {columns[1]} TEXT NOT NULL
                     )"""
-
-    # Create table
     cursor.execute(create_table_sql)
 
     # insert
@@ -140,11 +144,15 @@ def save_set(connection: sqlite3.Connection, table_name: str, rows: Set, column:
     :return: success status
     """
     cursor = connection.cursor()
+
+    # Drop the table first
+    drop_table_sql = f"DROP TABLE IF EXISTS {table_name}"
+    cursor.execute(drop_table_sql)
+
+    # Create table
     create_table_sql = f"""CREATE TABLE {table_name} (
                             {column} TEXT PRIMARY KEY
                     )"""
-
-    # Create table
     cursor.execute(create_table_sql)
 
     # insert

pyard/pyard.py

@@ -41,7 +41,8 @@ class ARD(object):
 
     def __init__(self, imgt_version: str = 'Latest',
                  remove_invalid: bool = True,
-                 data_dir: str = None):
+                 data_dir: str = None,
+                 refresh_mac: bool = False) -> None:
         """
         ARD will load valid alleles, xx codes and MAC mappings for the given
         version of IMGT database, downloading and generating the database if
@@ -57,7 +58,7 @@ def __init__(self, imgt_version: str = 'Latest',
         self.db_connection = db.create_db_connection(data_dir, imgt_version)
 
         # Load MAC codes
-        generate_mac_codes(self.db_connection)
+        generate_mac_codes(self.db_connection, refresh_mac)
         # Load Alleles and XX Codes
         self.valid_alleles, self.xx_codes = generate_alleles_and_xx_codes(self.db_connection, imgt_version)
         # Load ARS mappings