def get_db_connection(filename):
"""Make a connection to an SQLite database."""
try:
conn = sqlite3.connect(filename)
except:
logger.error("Failed to connect to SQLite db %s (exiting)" % filename)
logger.error(last_exception())
sys.exit(1)
return conn
def get_db_connection(filename):
"""Make a connection to an SQLite database."""
try:
conn = sqlite3.connect(filename)
except:
logger.error("Failed to connect to SQLite db %s (exiting)" % filename)
logger.error(last_exception())
sys.exit(1)
return conn
def create_db_tables(conn):
"""Create the tables for the SQLite3 database."""
# SQL for each table creation
sql_snp = """ DROP TABLE IF EXISTS snp_location;
CREATE TABLE snp_location (snp_id TEXT,
hg_version TEXT,
chromosome TEXT,
position INTEGER,
PRIMARY KEY(snp_id, hg_version));
"""
sql_gtype = """ DROP TABLE IF EXISTS genotypes;
CREATE TABLE genotypes
(snp_id TEXT NOT NULL,
genotype TEXT NOT NULL,
PRIMARY KEY(snp_id, genotype),
FOREIGN KEY(snp_id) REFERENCES snp(snp_id)
ON DELETE CASCADE);
"""
sql_person = """ DROP TABLE IF EXISTS person;
CREATE TABLE person
(person_id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT)
"""
sql_person_gtype = """ DROP TABLE IF EXISTS person_gtype;
CREATE TABLE person_gtype
(person_id INTEGER NOT NULL,
snp_id INTEGER NOT NULL,
genotype TEXT NOT NULL,
PRIMARY KEY(person_id, snp_id, genotype),
FOREIGN KEY(person_id)
REFERENCES person(person_id)
ON DELETE CASCADE,
FOREIGN KEY(snp_id, genotype)
REFERENCES snp(snp_id, genotype)
ON DELETE CASCADE);
"""
# Create each table in turn
for tname, sql in [("snp_location", sql_snp),
("genotype", sql_gtype),
("person", sql_person),
("person_gtype", sql_person_gtype)]:
with conn:
cur = conn.cursor()
try:
logger.info("Creating table %s" % tname)
cur.executescript(sql)
except:
logger.error("Could not create table % (exiting)" % tname)
logger.error(last_exception())
sys.exit(1)
def create_db_tables(conn):
"""Create the tables for the SQLite3 database."""
# SQL for each table creation
sql_snp = """ DROP TABLE IF EXISTS snp_location;
CREATE TABLE snp_location (snp_id TEXT,
hg_version TEXT,
chromosome TEXT,
position INTEGER,
PRIMARY KEY(snp_id, hg_version));
"""
sql_gtype = """ DROP TABLE IF EXISTS genotypes;
CREATE TABLE genotypes
(snp_id TEXT NOT NULL,
genotype TEXT NOT NULL,
PRIMARY KEY(snp_id, genotype),
FOREIGN KEY(snp_id) REFERENCES snp(snp_id)
ON DELETE CASCADE);
"""
sql_person = """ DROP TABLE IF EXISTS person;
CREATE TABLE person
(person_id INTEGER PRIMARY KEY AUTOINCREMENT,
name TEXT)
"""
sql_person_gtype = """ DROP TABLE IF EXISTS person_gtype;
CREATE TABLE person_gtype
(person_id INTEGER NOT NULL,
snp_id INTEGER NOT NULL,
genotype TEXT NOT NULL,
PRIMARY KEY(person_id, snp_id, genotype),
FOREIGN KEY(person_id)
REFERENCES person(person_id)
ON DELETE CASCADE,
FOREIGN KEY(snp_id, genotype)
REFERENCES snp(snp_id, genotype)
ON DELETE CASCADE);
"""
# Create each table in turn
for tname, sql in [("snp_location", sql_snp), ("genotype", sql_gtype),
("person", sql_person),
("person_gtype", sql_person_gtype)]:
with conn:
cur = conn.cursor()
try:
logger.info("Creating table %s" % tname)
cur.executescript(sql)
except:
logger.error("Could not create table % (exiting)" % tname)
logger.error(last_exception())
sys.exit(1)
def create_db_views(conn):
"""Create views on the database."""
sql_snpcount = """ DROP VIEW IF EXISTS snp_counts;
CREATE VIEW snp_counts
AS SELECT snp_id, COUNT(*) as count
FROM genotypes GROUP BY snp_id;
"""
# Create each view in turn
for tname, sql in [("snpcount", sql_snpcount)]:
with conn:
cur = conn.cursor()
try:
logger.info("Creating view %s" % tname)
cur.executescript(sql)
except:
logger.error("Could not create view % (exiting)" % tname)
logger.error(last_exception())
sys.exit(1)
def create_db_views(conn):
"""Create views on the database."""
sql_snpcount = """ DROP VIEW IF EXISTS snp_counts;
CREATE VIEW snp_counts
AS SELECT snp_id, COUNT(*) as count
FROM genotypes GROUP BY snp_id;
"""
# Create each view in turn
for tname, sql in [("snpcount", sql_snpcount)]:
with conn:
cur = conn.cursor()
try:
logger.info("Creating view %s" % tname)
cur.executescript(sql)
except:
logger.error("Could not create view % (exiting)" % tname)
logger.error(last_exception())
sys.exit(1)
def populate_db(name, snpfile, asm, conn):
"""Using the database with the connection in conn, add data for the named
person from the passed snpfile.
"""
# Get filehandle
try:
fh = open(snpfile, 'rU')
except:
logger.error("Could not open SNP file %s" % snpfile)
logger.error(last_exception())
sys.exit(1)
# Load data into database
logger.info("Processing %s" % snpfile)
with fh as snpfh:
with conn:
cur = conn.cursor()
# Populate the person table. At the mom
sql = "INSERT INTO person(name) VALUES (?)"
cur.execute(sql, (name, ))
person_id = cur.lastrowid
# Parse the SNP file
snpreader = csv.reader(snpfh, delimiter='\t')
try:
for row in snpreader:
if not row[0].startswith('#'): # ignore comments
rsid, chrm, pos, gt = tuple(row)
pos = int(pos)
# Add snp location
try:
sql = "INSERT INTO snp_location(snp_id, " +\
"hg_version, chromosome, " +\
"position) VALUES (?, ?, ?, ?)"
cur.execute(sql, (rsid, asm, chrm, pos))
except sqlite3.IntegrityError:
# This will throw an error if the SNP location (on
# this HG build) is already found in the db
logger.warning("SNP %s position " % rsid +
"on HG assembly build " +
"%s already present" % asm)
# Insert genotype
try:
sql = "INSERT INTO genotypes(snp_id, genotype) " +\
"VALUES (?, ?)"
cur.execute(sql, (rsid, gt))
except sqlite3.IntegrityError:
# This will throw an error if the genotype exists
# in the db
logger.warning("Genotype %s already " % gt +
"present for SNP %s" % rsid)
# Link individual to genotype
try:
sql = "INSERT INTO person_gtype(person_id, " +\
"snp_id, genotype) VALUES (?, ?, ?)"
cur.execute(sql, (person_id, rsid, gt))
except:
logger.error("Problem populating database at " +
"row %s (exiting)" % row)
sys.exit(1)
except:
logger.error("Problem parsing SNP file %s" % snpfile)
logger.error(last_exception())
sys.exit(1)
def populate_db(name, snpfile, asm, conn):
"""Using the database with the connection in conn, add data for the named
person from the passed snpfile.
"""
# Get filehandle
try:
fh = open(snpfile, 'rU')
except:
logger.error("Could not open SNP file %s" % snpfile)
logger.error(last_exception())
sys.exit(1)
# Load data into database
logger.info("Processing %s" % snpfile)
with fh as snpfh:
with conn:
cur = conn.cursor()
# Populate the person table. At the mom
sql = "INSERT INTO person(name) VALUES (?)"
cur.execute(sql, (name, ))
person_id = cur.lastrowid
# Parse the SNP file
snpreader = csv.reader(snpfh, delimiter='\t')
try:
for row in snpreader:
if not row[0].startswith('#'): # ignore comments
rsid, chrm, pos, gt = tuple(row)
pos = int(pos)
# Add snp location
try:
sql = "INSERT INTO snp_location(snp_id, " +\
"hg_version, chromosome, " +\
"position) VALUES (?, ?, ?, ?)"
cur.execute(sql, (rsid, asm, chrm, pos))
except sqlite3.IntegrityError:
# This will throw an error if the SNP location (on
# this HG build) is already found in the db
logger.warning("SNP %s position " % rsid +
"on HG assembly build " +
"%s already present" % asm)
# Insert genotype
try:
sql = "INSERT INTO genotypes(snp_id, genotype) " +\
"VALUES (?, ?)"
cur.execute(sql, (rsid, gt))
except sqlite3.IntegrityError:
# This will throw an error if the genotype exists
# in the db
logger.warning("Genotype %s already " % gt +
"present for SNP %s" % rsid)
# Link individual to genotype
try:
sql = "INSERT INTO person_gtype(person_id, " +\
"snp_id, genotype) VALUES (?, ?, ?)"
cur.execute(sql, (person_id, rsid, gt))
except:
logger.error("Problem populating database at " +
"row %s (exiting)" % row)
sys.exit(1)
except:
logger.error("Problem parsing SNP file %s" % snpfile)
logger.error(last_exception())
sys.exit(1)