def test_remove_common_words(self):
""" Test removing common words """
d = "This is the sentence of words to process"
words = NRDatabaseFilter.remove_common_words(d.split())
self.assertTrue(not "the" in words)
self.assertTrue(not "of" in words)
self.assertTrue("sentence" in words)
self.assertTrue("This" in words)
def test_filtering(self):
""" test the filtering of the NR database """
f = open(os.path.join(self.datadir, "coglist13962_10-oct-2012.txt"), "rU") # rU read with universal new line
reader = csv.reader(f, delimiter="\t")
reader.next() # discard title line
i = 0
cogs_names = []
cogs_ids = []
for row in reader:
cogs_ids.append(row[0])
cogs_names.append(row[1].lower())
i += 1
if i > 5:
break
f.close()
log.debug("Filtering COGS %s", cogs_ids)
log.debug("Descriptions: %s", cogs_names)
fn_database = os.path.join(self.datadir, "mini_nr", "nr_test2")
dbfilter = NRDatabaseFilter.NRDatabaseFilter(fn_database)
dbfilter.set_ids(cogs_ids)
dbfilter.set_descriptions(cogs_names)
dbfilter.do_filtering(overwrite=True)
for cog_id, name in zip(cogs_ids, cogs_names):
fn = dbfilter.get_database_name(cog_id)
if os.path.exists(fn):
# check that the file contains only sequences that
# match the description
keywords = name.lower().split()
for seq_record in SeqIO.parse(fn, "fasta"):
words = seq_record.description.lower().split(" ")
self.assertTrue(
NRDatabaseFilter.contains_all_keywords(words, keywords),
"There is a sequence in file {0} that does not have "
"the right kewords {1}".format(fn, str(keywords)),
)
map(os.remove, dbfilter.get_database_files_created())
def test_contains_all_keywords(self):
""" Test contains all keywords """
words = "tRNA amylase somethig protein".split()
keywords = "amylase protein".split()
self.assertTrue(NRDatabaseFilter.contains_all_keywords(words, keywords))
