# This code is part of the Biopython distribution and governed by its # license. Please see the LICENSE file that should have been included # as part of this package. """Tests for dealing with storage of biopython objects in a relational db. """ from __future__ import print_function import os import platform import unittest import tempfile import time from Bio._py3k import StringIO from Bio._py3k import zip from Bio._py3k import basestring # Hide annoying warnings from things like bonds in GenBank features, # or PostgreSQL schema rules. TODO - test these warnings are raised! import warnings from Bio import BiopythonWarning # local stuff from Bio import MissingExternalDependencyError from Bio.Seq import Seq, MutableSeq from Bio.SeqFeature import SeqFeature from Bio import Alphabet from Bio import SeqIO from Bio.SeqRecord import SeqRecord from BioSQL import BioSeqDatabase from BioSQL import BioSeq from seq_tests_common import compare_record, compare_records if __name__ == "__main__": raise RuntimeError("Call this via test_BioSQL_*.py not directly") global DBDRIVER, DBTYPE, DBHOST, DBUSER, DBPASSWD, TESTDB, DBSCHEMA, SQL_FILE global SYSTEM SYSTEM = platform.system() def temp_db_filename(): # In memory SQLite does not work with current test structure since the tests # expect databases to be retained between individual tests. # TESTDB = ':memory:' # Instead, we use (if we can) /dev/shm try: h, test_db_fname = tempfile.mkstemp("_BioSQL.db", dir='/dev/shm') except OSError: # We can't use /dev/shm h, test_db_fname = tempfile.mkstemp("_BioSQL.db") os.close(h) return test_db_fname def check_config(dbdriver, dbtype, dbhost, dbuser, dbpasswd, testdb): global DBDRIVER, DBTYPE, DBHOST, DBUSER, DBPASSWD, TESTDB, DBSCHEMA global SYSTEM, SQL_FILE DBDRIVER = dbdriver DBTYPE = dbtype DBHOST = dbhost DBUSER = dbuser DBPASSWD = dbpasswd TESTDB = testdb #Check the database driver is installed: if SYSTEM == "Java": try: if DBDRIVER in ["MySQLdb"]: import com.mysql.jdbc.Driver elif DBDRIVER in ["psycopg2"]: import org.postgresql.Driver except ImportError: message = "Install the JDBC driver for %s to use BioSQL " % DBTYPE raise MissingExternalDependencyError(message) else: try: __import__(DBDRIVER) except ImportError: message = "Install %s if you want to use %s with BioSQL " % (DBDRIVER, DBTYPE) raise MissingExternalDependencyError(message) try: if DBDRIVER in ["sqlite3"]: server = BioSeqDatabase.open_database(driver=DBDRIVER, db=TESTDB) else: server = BioSeqDatabase.open_database(driver=DBDRIVER, user=DBUSER, passwd=DBPASSWD, host=DBHOST) server.close() del server except Exception as e: message = "Connection failed, check settings if you plan to use BioSQL: %s" % str(e) raise MissingExternalDependencyError(message) DBSCHEMA = "biosqldb-" + DBTYPE + ".sql" SQL_FILE = os.path.join(os.getcwd(), "BioSQL", DBSCHEMA) if not os.path.isfile(SQL_FILE): message = "Missing SQL schema file: %s" % SQL_FILE raise MissingExternalDependencyError(message) def _do_db_create(): """Do the actual work of database creation. Relevant for MySQL and PostgreSQL """ # first open a connection to create the database server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST) if DBDRIVER == "pgdb": # The pgdb postgres driver does not support autocommit, so here we # commit the current transaction so that 'drop database' query will # be outside a transaction block server.adaptor.cursor.execute("COMMIT") else: # Auto-commit: postgresql cannot drop database in a transaction try: server.adaptor.autocommit() except AttributeError: pass # drop anything in the database try: # with Postgres, can get errors about database still being used and # not able to be dropped. Wait briefly to be sure previous tests are # done with it. time.sleep(1) sql = r"DROP DATABASE " + TESTDB server.adaptor.cursor.execute(sql, ()) except (server.module.OperationalError, server.module.Error, server.module.DatabaseError) as e: # the database doesn't exist pass except (server.module.IntegrityError, server.module.ProgrammingError) as e: # ditto--perhaps if str(e).find('database "%s" does not exist' % TESTDB) == -1: server.close() raise # create a new database sql = r"CREATE DATABASE " + TESTDB server.adaptor.execute(sql, ()) server.close() def create_database(): """Delete any existing BioSQL test database, then (re)create an empty BioSQL database.""" if DBDRIVER in ["sqlite3"]: global TESTDB if os.path.exists(TESTDB): try: os.remove(TESTDB) except: time.sleep(1) try: os.remove(TESTDB) except: # Seen this with PyPy 2.1 (and older) on Windows - # which suggests an open handle still exists? print("Could not remove %r" % TESTDB) pass # Now pick a new filename - just in case there is a stale handle # (which might be happening under Windows...) TESTDB = temp_db_filename() else: _do_db_create() # now open a connection to load the database server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) try: server.load_database_sql(SQL_FILE) server.commit() server.close() except: # Failed, but must close the handle... server.close() raise def destroy_database(): """Delete any temporary BioSQL sqlite3 database files.""" if DBDRIVER in ["sqlite3"]: if os.path.exists(TESTDB): os.remove(TESTDB) def load_database(gb_filename_or_handle): """Load a GenBank file into a new BioSQL database. This is useful for running tests against a newly created database. """ create_database() # now open a connection to load the database db_name = "biosql-test" server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) db = server.new_database(db_name) # get the GenBank file we are going to put into it iterator = SeqIO.parse(gb_filename_or_handle, "gb") # finally put it in the database count = db.load(iterator) server.commit() server.close() return count class ReadTest(unittest.TestCase): """Test reading a database from an already built database. """ loaded_db = 0 def setUp(self): """Connect to and load up the database. """ load_database("GenBank/cor6_6.gb") self.server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) self.db = self.server["biosql-test"] def tearDown(self): self.server.close() destroy_database() del self.db del self.server def test_server(self): """Check BioSeqDatabase methods""" server = self.server self.assertTrue("biosql-test" in server) self.assertEqual(1, len(server)) self.assertEqual(["biosql-test"], list(server.keys())) #Check we can delete the namespace... del server["biosql-test"] self.assertEqual(0, len(server)) try: del server["non-existant-name"] assert False, "Should have raised KeyError" except KeyError: pass def test_get_db_items(self): """Check list, keys, length etc""" db = self.db items = list(db.values()) keys = list(db) l = len(items) self.assertEqual(l, len(db)) self.assertEqual(l, len(list(db.items()))) self.assertEqual(l, len(list(db))) self.assertEqual(l, len(list(db.values()))) for (k1, r1), (k2, r2) in zip(zip(keys, items), db.items()): self.assertEqual(k1, k2) self.assertEqual(r1.id, r2.id) for k in keys: del db[k] self.assertEqual(0, len(db)) try: del db["non-existant-name"] assert False, "Should have raised KeyError" except KeyError: pass def test_lookup_items(self): """Test retrieval of items using various ids. """ self.db.lookup(accession = "X62281") try: self.db.lookup(accession = "Not real") raise AssertionError("No problem on fake id retrieval") except IndexError: pass self.db.lookup(display_id = "ATKIN2") try: self.db.lookup(display_id = "Not real") raise AssertionError("No problem on fake id retrieval") except IndexError: pass # primary id retrieval self.db.lookup(primary_id = "16353") try: self.db.lookup(primary_id = "Not Real") raise AssertionError("No problem on fake primary id retrieval") except IndexError: pass class SeqInterfaceTest(unittest.TestCase): """Make sure the BioSQL objects implement the expected biopython interfaces """ def setUp(self): """Load a database. """ load_database("GenBank/cor6_6.gb") self.server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) self.db = self.server["biosql-test"] self.item = self.db.lookup(accession = "X62281") def tearDown(self): self.server.close() destroy_database() del self.db del self.item del self.server def test_seq_record(self): """Make sure SeqRecords from BioSQL implement the right interface. """ test_record = self.item self.assertTrue(isinstance(test_record.seq, BioSeq.DBSeq)) self.assertEqual(test_record.id, "X62281.1", test_record.id) self.assertEqual(test_record.name, "ATKIN2") self.assertEqual(test_record.description, "A.thaliana kin2 gene.") self.assertTrue(hasattr(test_record, 'annotations')) # XXX should do something with annotations once they are like # a dictionary for feature in test_record.features: self.assertTrue(isinstance(feature, SeqFeature)) self.assertTrue(isinstance(str(test_record), basestring)) # shouldn't cause any errors! def test_seq(self): """Make sure Seqs from BioSQL implement the right interface. """ test_seq = self.item.seq alphabet = test_seq.alphabet self.assertTrue(isinstance(alphabet, Alphabet.Alphabet)) data = test_seq.data self.assertEqual(type(data), type("")) string_rep = str(test_seq) self.assertEqual(string_rep, str(test_seq)) # check __str__ too self.assertEqual(type(string_rep), type("")) self.assertEqual(len(test_seq), 880) def test_convert(self): """Check can turn a DBSeq object into a Seq or MutableSeq.""" test_seq = self.item.seq other = test_seq.toseq() self.assertEqual(str(test_seq), str(other)) self.assertEqual(test_seq.alphabet, other.alphabet) self.assertTrue(isinstance(other, Seq)) other = test_seq.tomutable() self.assertEqual(str(test_seq), str(other)) self.assertEqual(test_seq.alphabet, other.alphabet) self.assertTrue(isinstance(other, MutableSeq)) def test_addition(self): """Check can add DBSeq objects together.""" test_seq = self.item.seq for other in [Seq("ACGT", test_seq.alphabet), MutableSeq("ACGT", test_seq.alphabet), "ACGT", test_seq]: test = test_seq + other self.assertEqual(str(test), str(test_seq) + str(other)) self.assertTrue(isinstance(test, Seq)) test = other + test_seq self.assertEqual(str(test), str(other) + str(test_seq)) def test_seq_slicing(self): """Check that slices of sequences are retrieved properly. """ test_seq = self.item.seq new_seq = test_seq[:10] self.assertTrue(isinstance(new_seq, BioSeq.DBSeq)) # simple slicing self.assertEqual(str(test_seq[:5]), 'ATTTG') self.assertEqual(str(test_seq[0:5]), 'ATTTG') self.assertEqual(str(test_seq[2:3]), 'T') self.assertEqual(str(test_seq[2:4]), 'TT') self.assertEqual(str(test_seq[870:]), 'TTGAATTATA') # getting more fancy self.assertEqual(test_seq[-1], 'A') self.assertEqual(test_seq[1], 'T') self.assertEqual(str(test_seq[-10:][5:]), "TTATA") self.assertEqual(str(test_seq[-10:][5:]), "TTATA") def test_seq_features(self): """Check SeqFeatures of a sequence. """ test_features = self.item.features cds_feature = test_features[6] self.assertEqual(cds_feature.type, "CDS") self.assertEqual(str(cds_feature.location), "join{[103:160](+), [319:390](+), [503:579](+)}") for sub_feature in cds_feature._sub_features: self.assertEqual(sub_feature.type, "CDS") #self.assertEqual(sub_feature.location_operator, "join") try: self.assertEqual(cds_feature.qualifiers["gene"], ["kin2"]) self.assertEqual(cds_feature.qualifiers["protein_id"], ["CAA44171.1"]) self.assertEqual(cds_feature.qualifiers["codon_start"], ["1"]) except KeyError: raise KeyError("Missing expected entries, have %s" % repr(cds_feature.qualifiers)) self.assertTrue("db_xref" in cds_feature.qualifiers) multi_ann = cds_feature.qualifiers["db_xref"] self.assertEqual(len(multi_ann), 2) self.assertTrue("GI:16354" in multi_ann) self.assertTrue("SWISS-PROT:P31169" in multi_ann) class LoaderTest(unittest.TestCase): """Load a database from a GenBank file. """ def setUp(self): # create TESTDB create_database() # load the database db_name = "biosql-test" self.server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) # remove the database if it already exists try: self.server[db_name] self.server.remove_database(db_name) except KeyError: pass self.db = self.server.new_database(db_name) # get the GenBank file we are going to put into it self.iterator = SeqIO.parse("GenBank/cor6_6.gb", "gb") def tearDown(self): self.server.close() destroy_database() del self.db del self.server def test_load_database(self): """Load SeqRecord objects into a BioSQL database. """ self.db.load(self.iterator) # do some simple tests to make sure we actually loaded the right # thing. More advanced tests in a different module. items = list(self.db.values()) self.assertEqual(len(items), 6) self.assertEqual(len(self.db), 6) item_names = [] item_ids = [] for item in items: item_names.append(item.name) item_ids.append(item.id) item_names.sort() item_ids.sort() self.assertEqual(item_names, ['AF297471', 'ARU237582', 'ATCOR66M', 'ATKIN2', 'BNAKINI', 'BRRBIF72']) self.assertEqual(item_ids, ['AF297471.1', 'AJ237582.1', 'L31939.1', 'M81224.1', 'X55053.1', 'X62281.1']) class DupLoadTest(unittest.TestCase): """Check a few duplicate conditions fail.""" def setUp(self): #drop any old database and create a new one: create_database() #connect to new database: self.server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) #Create new namespace within new empty database: self.db = self.server.new_database("biosql-test") def tearDown(self): self.server.rollback() self.server.close() destroy_database() del self.db del self.server def test_duplicate_load(self): """Make sure can't import a single record twice (in one go).""" record = SeqRecord(Seq("ATGCTATGACTAT", Alphabet.generic_dna), id="Test1") try: count = self.db.load([record, record]) except Exception as err: #Good! #Note we don't do a specific exception handler because the #exception class will depend on which DB back end is in use. self.assertTrue(err.__class__.__name__ in ["IntegrityError", "AttributeError", "OperationalError"], err.__class__.__name__) return raise Exception("Should have failed! Loaded %i records" % count) def test_duplicate_load2(self): """Make sure can't import a single record twice (in steps).""" record = SeqRecord(Seq("ATGCTATGACTAT", Alphabet.generic_dna), id="Test2") count = self.db.load([record]) self.assertEqual(count, 1) try: count = self.db.load([record]) except Exception as err: #Good! self.assertTrue(err.__class__.__name__ in ["IntegrityError", "AttributeError"], err.__class__.__name__) return raise Exception("Should have failed! Loaded %i records" % count) def test_duplicate_id_load(self): """Make sure can't import records with same ID (in one go).""" record1 = SeqRecord(Seq("ATGCTATGACTAT", Alphabet.generic_dna), id="TestA") record2 = SeqRecord(Seq("GGGATGCGACTAT", Alphabet.generic_dna), id="TestA") try: count = self.db.load([record1, record2]) except Exception as err: #Good! self.assertTrue(err.__class__.__name__ in ["IntegrityError", "AttributeError"], err.__class__.__name__) return raise Exception("Should have failed! Loaded %i records" % count) class ClosedLoopTest(unittest.TestCase): """Test file -> BioSQL -> file.""" #NOTE - For speed I don't bother to create a new database each time, #simple a new unique namespace is used for each test. def test_NC_005816(self): """GenBank file to BioSQL and back to a GenBank file, NC_005816.""" self.loop("GenBank/NC_005816.gb", "gb") def test_NC_000932(self): """GenBank file to BioSQL and back to a GenBank file, NC_000932.""" self.loop("GenBank/NC_000932.gb", "gb") def test_NT_019265(self): """GenBank file to BioSQL and back to a GenBank file, NT_019265.""" self.loop("GenBank/NT_019265.gb", "gb") def test_protein_refseq2(self): """GenBank file to BioSQL and back to a GenBank file, protein_refseq2.""" self.loop("GenBank/protein_refseq2.gb", "gb") def test_no_ref(self): """GenBank file to BioSQL and back to a GenBank file, noref.""" self.loop("GenBank/noref.gb", "gb") def test_one_of(self): """GenBank file to BioSQL and back to a GenBank file, one_of.""" self.loop("GenBank/one_of.gb", "gb") def test_cor6_6(self): """GenBank file to BioSQL and back to a GenBank file, cor6_6.""" self.loop("GenBank/cor6_6.gb", "gb") def test_arab1(self): """GenBank file to BioSQL and back to a GenBank file, arab1.""" self.loop("GenBank/arab1.gb", "gb") def loop(self, filename, format): original_records = list(SeqIO.parse(filename, format)) # now open a connection to load the database server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) db_name = "test_loop_%s" % filename # new namespace! db = server.new_database(db_name) count = db.load(original_records) self.assertEqual(count, len(original_records)) server.commit() #Now read them back... biosql_records = [db.lookup(name=rec.name) for rec in original_records] #And check they agree self.assertTrue(compare_records(original_records, biosql_records)) #Now write to a handle... handle = StringIO() SeqIO.write(biosql_records, handle, "gb") #Now read them back... handle.seek(0) new_records = list(SeqIO.parse(handle, "gb")) #And check they still agree self.assertEqual(len(new_records), len(original_records)) for old, new in zip(original_records, new_records): #TODO - remove this hack because we don't yet write these (yet): for key in ["comment", "references", "db_source"]: if key in old.annotations and key not in new.annotations: del old.annotations[key] self.assertTrue(compare_record(old, new)) #Done handle.close() server.close() class TransferTest(unittest.TestCase): """Test file -> BioSQL, BioSQL -> BioSQL.""" #NOTE - For speed I don't bother to create a new database each time, #simple a new unique namespace is used for each test. def setUp(self): create_database() def test_NC_005816(self): """GenBank file to BioSQL, then again to a new namespace, NC_005816.""" self.trans("GenBank/NC_005816.gb", "gb") def test_NC_000932(self): """GenBank file to BioSQL, then again to a new namespace, NC_000932.""" self.trans("GenBank/NC_000932.gb", "gb") def test_NT_019265(self): """GenBank file to BioSQL, then again to a new namespace, NT_019265.""" self.trans("GenBank/NT_019265.gb", "gb") def test_protein_refseq2(self): """GenBank file to BioSQL, then again to a new namespace, protein_refseq2.""" self.trans("GenBank/protein_refseq2.gb", "gb") def test_no_ref(self): """GenBank file to BioSQL, then again to a new namespace, noref.""" self.trans("GenBank/noref.gb", "gb") def test_one_of(self): """GenBank file to BioSQL, then again to a new namespace, one_of.""" self.trans("GenBank/one_of.gb", "gb") def test_cor6_6(self): """GenBank file to BioSQL, then again to a new namespace, cor6_6.""" self.trans("GenBank/cor6_6.gb", "gb") def test_arab1(self): """GenBank file to BioSQL, then again to a new namespace, arab1.""" self.trans("GenBank/arab1.gb", "gb") def trans(self, filename, format): original_records = list(SeqIO.parse(filename, format)) # now open a connection to load the database server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) db_name = "test_trans1_%s" % filename # new namespace! db = server.new_database(db_name) count = db.load(original_records) self.assertEqual(count, len(original_records)) server.commit() #Now read them back... biosql_records = [db.lookup(name=rec.name) for rec in original_records] #And check they agree self.assertTrue(compare_records(original_records, biosql_records)) #Now write to a second name space... db_name = "test_trans2_%s" % filename # new namespace! db = server.new_database(db_name) count = db.load(biosql_records) self.assertEqual(count, len(original_records)) #Now read them back again, biosql_records2 = [db.lookup(name=rec.name) for rec in original_records] #And check they also agree self.assertTrue(compare_records(original_records, biosql_records2)) #Done server.close() def tearDown(self): destroy_database() class InDepthLoadTest(unittest.TestCase): """Make sure we are loading and retreiving in a semi-lossless fashion. """ def setUp(self): gb_file = os.path.join(os.getcwd(), "GenBank", "cor6_6.gb") load_database(gb_file) self.server = BioSeqDatabase.open_database(driver = DBDRIVER, user = DBUSER, passwd = DBPASSWD, host = DBHOST, db = TESTDB) self.db = self.server["biosql-test"] def tearDown(self): self.server.close() destroy_database() del self.db del self.server def test_transfer(self): """Make sure can load record into another namespace.""" #Should be in database already... db_record = self.db.lookup(accession = "X55053") #Make a new namespace db2 = self.server.new_database("biosql-test-alt") #Should be able to load this DBSeqRecord there... count = db2.load([db_record]) self.assertEqual(count, 1) def test_reload(self): """Make sure can't reimport existing records.""" gb_file = os.path.join(os.getcwd(), "GenBank", "cor6_6.gb") gb_handle = open(gb_file, "r") record = next(SeqIO.parse(gb_handle, "gb")) gb_handle.close() #Should be in database already... db_record = self.db.lookup(accession = "X55053") self.assertEqual(db_record.id, record.id) self.assertEqual(db_record.name, record.name) self.assertEqual(db_record.description, record.description) self.assertEqual(str(db_record.seq), str(record.seq)) #Good... now try reloading it! try: count = self.db.load([record]) except Exception as err: #Good! self.assertTrue(err.__class__.__name__ in ["IntegrityError", "AttributeError"], err.__class__.__name__) return raise Exception("Should have failed! Loaded %i records" % count) def test_record_loading(self): """Make sure all records are correctly loaded. """ test_record = self.db.lookup(accession = "X55053") self.assertEqual(test_record.name, "ATCOR66M") self.assertEqual(test_record.id, "X55053.1") self.assertEqual(test_record.description, "A.thaliana cor6.6 mRNA.") self.assertTrue(isinstance(test_record.seq.alphabet, Alphabet.DNAAlphabet)) self.assertEqual(str(test_record.seq[:10]), 'AACAAAACAC') test_record = self.db.lookup(accession = "X62281") self.assertEqual(test_record.name, "ATKIN2") self.assertEqual(test_record.id, "X62281.1") self.assertEqual(test_record.description, "A.thaliana kin2 gene.") self.assertTrue(isinstance(test_record.seq.alphabet, Alphabet.DNAAlphabet)) self.assertEqual(str(test_record.seq[:10]), 'ATTTGGCCTA') def test_seq_feature(self): """In depth check that SeqFeatures are transmitted through the db. """ test_record = self.db.lookup(accession = "AJ237582") features = test_record.features self.assertEqual(len(features), 7) # test single locations test_feature = features[0] self.assertEqual(test_feature.type, "source") self.assertEqual(str(test_feature.location), "[0:206](+)") self.assertEqual(len(test_feature.qualifiers), 3) self.assertEqual(test_feature.qualifiers["country"], ["Russia:Bashkortostan"]) self.assertEqual(test_feature.qualifiers["organism"], ["Armoracia rusticana"]) self.assertEqual(test_feature.qualifiers["db_xref"], ["taxon:3704"]) # test split locations test_feature = features[4] self.assertEqual(test_feature.type, "CDS") self.assertEqual(str(test_feature.location), "join{[0:48](+), [142:206](+)}") self.assertEqual(len(test_feature._sub_features), 2) self.assertEqual(str(test_feature._sub_features[0].location), "[0:48](+)") self.assertEqual(test_feature._sub_features[0].type, "CDS") #self.assertEqual(test_feature._sub_features[0].location_operator, "join") self.assertEqual(str(test_feature._sub_features[1].location), "[142:206](+)") self.assertEqual(test_feature._sub_features[1].type, "CDS") #self.assertEqual(test_feature._sub_features[1].location_operator, "join") self.assertEqual(len(test_feature.qualifiers), 6) self.assertEqual(test_feature.qualifiers["gene"], ["csp14"]) self.assertEqual(test_feature.qualifiers["codon_start"], ["2"]) self.assertEqual(test_feature.qualifiers["product"], ["cold shock protein"]) self.assertEqual(test_feature.qualifiers["protein_id"], ["CAB39890.1"]) self.assertEqual(test_feature.qualifiers["db_xref"], ["GI:4538893"]) self.assertEqual(test_feature.qualifiers["translation"], ["DKAKDAAAAAGASAQQAGKNISDAAAGGVNFVKEKTG"]) # test passing strand information # XXX We should be testing complement as well test_record = self.db.lookup(accession = "AJ237582") test_feature = test_record.features[4] # DNA, no complement self.assertEqual(test_feature.strand, 1) for sub_feature in test_feature._sub_features: self.assertEqual(sub_feature.strand, 1) test_record = self.db.lookup(accession = "X55053") test_feature = test_record.features[0] # mRNA, so really cDNA, so the strand should be 1 (not complemented) self.assertEqual(test_feature.strand, 1) ##################################################################### class AutoSeqIOTests(unittest.TestCase): """Test SeqIO and BioSQL together.""" server = None db = None def setUp(self): """Connect to the database.""" db_name = "biosql-test-seqio" server = BioSeqDatabase.open_database(driver=DBDRIVER, user=DBUSER, passwd=DBPASSWD, host=DBHOST, db=TESTDB) self.server = server if db_name not in server: self.db = server.new_database(db_name) server.commit() self.db = self.server[db_name] def tearDown(self): if self.db: del self.db if self.server: self.server.close() del self.server def check(self, t_format, t_filename, t_count=1): db = self.db iterator = SeqIO.parse(t_filename, t_format) count = db.load(iterator) assert count == t_count self.server.commit() iterator = SeqIO.parse(t_filename, t_format) for record in iterator: #print(" - %s, %s" % (checksum_summary(record), record.id)) key = record.name #print(" - Retrieving by name/display_id '%s'," % key) db_rec = db.lookup(name=key) compare_record(record, db_rec) db_rec = db.lookup(display_id=key) compare_record(record, db_rec) key = record.id if key.count(".") == 1 and key.split(".")[1].isdigit(): #print(" - Retrieving by version '%s'," % key) db_rec = db.lookup(version=key) compare_record(record, db_rec) if "accessions" in record.annotations: #Only expect FIRST accession to work! key = record.annotations["accessions"][0] assert key, "Blank accession in annotation %s" % repr(record.annotations) if key != record.id: #print(" - Retrieving by accession '%s'," % key) db_rec = db.lookup(accession=key) compare_record(record, db_rec) if "gi" in record.annotations: key = record.annotations['gi'] if key != record.id: #print(" - Retrieving by GI '%s'," % key) db_rec = db.lookup(primary_id=key) compare_record(record, db_rec) def test_SeqIO_loading(self): self.check('fasta', 'Fasta/lupine.nu') self.check('fasta', 'Fasta/elderberry.nu') self.check('fasta', 'Fasta/phlox.nu') self.check('fasta', 'Fasta/centaurea.nu') self.check('fasta', 'Fasta/wisteria.nu') self.check('fasta', 'Fasta/sweetpea.nu') self.check('fasta', 'Fasta/lavender.nu') self.check('fasta', 'Fasta/aster.pro') self.check('fasta', 'Fasta/loveliesbleeding.pro') self.check('fasta', 'Fasta/rose.pro') self.check('fasta', 'Fasta/rosemary.pro') self.check('fasta', 'Fasta/f001') self.check('fasta', 'Fasta/f002', 3) self.check('fasta', 'Fasta/fa01', 2) self.check('fasta', 'GFF/NC_001802.fna') self.check('fasta', 'GFF/multi.fna', 3) self.check('fasta', 'Registry/seqs.fasta', 2) self.check('swiss', 'SwissProt/sp001') self.check('swiss', 'SwissProt/sp002') self.check('swiss', 'SwissProt/sp003') self.check('swiss', 'SwissProt/sp004') self.check('swiss', 'SwissProt/sp005') self.check('swiss', 'SwissProt/sp006') self.check('swiss', 'SwissProt/sp007') self.check('swiss', 'SwissProt/sp008') self.check('swiss', 'SwissProt/sp009') self.check('swiss', 'SwissProt/sp010') self.check('swiss', 'SwissProt/sp011') self.check('swiss', 'SwissProt/sp012') self.check('swiss', 'SwissProt/sp013') self.check('swiss', 'SwissProt/sp014') self.check('swiss', 'SwissProt/sp015') self.check('swiss', 'SwissProt/sp016') self.check('swiss', 'Registry/EDD_RAT.dat') self.check('genbank', 'GenBank/noref.gb') self.check('genbank', 'GenBank/cor6_6.gb', 6) self.check('genbank', 'GenBank/iro.gb') self.check('genbank', 'GenBank/pri1.gb') self.check('genbank', 'GenBank/arab1.gb') with warnings.catch_warnings(): #BiopythonWarning: order location operators are not fully supported warnings.simplefilter("ignore", BiopythonWarning) self.check('genbank', 'GenBank/protein_refseq2.gb') self.check('genbank', 'GenBank/extra_keywords.gb') self.check('genbank', 'GenBank/one_of.gb') self.check('genbank', 'GenBank/NT_019265.gb') self.check('genbank', 'GenBank/origin_line.gb') self.check('genbank', 'GenBank/blank_seq.gb') with warnings.catch_warnings(): #BiopythonWarning: bond location operators are not fully supported warnings.simplefilter("ignore", BiopythonWarning) self.check('genbank', 'GenBank/dbsource_wrap.gb') self.check('genbank', 'GenBank/NC_005816.gb') self.check('genbank', 'GenBank/gbvrl1_start.seq', 3) self.check('genbank', 'GFF/NC_001422.gbk') self.check('embl', 'EMBL/TRBG361.embl') self.check('embl', 'EMBL/DD231055_edited.embl') self.check('embl', 'EMBL/SC10H5.embl') self.check('embl', 'EMBL/U87107.embl') self.assertEqual(len(self.db), 66)