def get_ppis(examples=10,
path_swissprot=config.PATH_SWISSPROT,
file_swissprot_proteins=config.FILE_SWISSPROT_PROTEINS,
url_swissprot=config.URL_SWISSPROT_PROTEINS,
path_reactome=config.PATH_REACTOME,
file_reactome_internal_edges=config.REACTOME_INTERACTIONS,
file_reactome_ppis=config.REACTOME_PPIS,
path_pathwaymatcher=config.PATH_TOOLS,
file_pathwaymatcher=config.FILE_PATHWAYMATCHER,
url_pathwaymatcher=config.URL_PATHWAYMATCHER):
"""Returns dictionary of lexicographical interactions: accessions --> accessions set"""
ppis = {}
if not os.path.exists(path_reactome + file_reactome_ppis):
create_pathwaymatcher_files(path_swissprot, file_swissprot_proteins,
url_swissprot, path_reactome,
file_reactome_internal_edges,
path_pathwaymatcher, file_pathwaymatcher,
url_pathwaymatcher)
print("Reading Reactome interactions...")
ppis = dictionaries.read_dictionary_one_to_set(
path_reactome,
file_reactome_internal_edges,
order_pairs=True,
col_indices=(0, 1),
ignore_header=True)
dictionaries.write_dictionary_one_to_set(ppis, path_reactome,
file_reactome_ppis)
else:
print("Reading Reactome unique interactions...")
ppis = dictionaries.read_dictionary_one_to_set(path_reactome,
file_reactome_ppis)
ppi_subset = {}
example = 0
if examples > 8000:
for key, values in ppis.items():
for value in values:
ppi_subset.setdefault(key.strip(), set()).add(value.strip())
example += 1
if example >= examples:
break
if example >= examples:
break
else:
random.seed(77)
keys = random.sample(list(ppis.keys()), int(examples))
for key in keys:
ppi_subset.setdefault(key.strip(), set()).add(
random.sample(ppis[key], 1)[0].strip())
print("Reactome interactions READY")
return ppi_subset
def test_read_dictionary_skip_header(self):
# Create trio file with headers
trios = [('Column1', 'Column2', 'Column3'), (1, 1, 2), (2, 3, 2),
(3, 4, 5)]
file_name = TestCase.id(self) + '_pairs.txt'
with open(file_name, 'w') as file:
for x, y, z in trios:
file.write(f"{x}\t{y}\t{z}\n")
# Execute target method
result = read_dictionary_one_to_set('',
file_name,
order_pairs=True,
col_indices=(1, 2),
ignore_header=True)
# Check headers are not taken as key, value pairs
self.assertNotIn('Column1',
result.keys(),
msg="Missing key in dictionary")
self.assertIn('1', result.keys(), msg="Missing key in dictionary")
self.assertIn('2', result.keys(), msg="Missing key in dictionary")
self.assertIn('4', result.keys(), msg="Missing key in dictionary")
# Remove precondition files
os.remove(file_name)
def test_read_dictionary_missing_two_columns(self):
"""With a one column file, request default columns 0 and 1, report error"""
# Create file with three columns, some not in lexicographic order
file_name = TestCase.id(self) + '_single_column.txt'
with open(file_name, 'w') as file:
for x in range(5):
file.write(f"{x}\n")
with self.assertRaises(
ValueError,
msg=
'Should raise an exception because needed columns of the file are missing.'
):
read_dictionary_one_to_set('', file_name, order_pairs=True)
os.remove(file_name)
def setUp(self):
self.letters = {chr(ord('A') + i): i for i in range(23)}
languages = [('C++', '.cpp'), ('Python', '.py'), ('JavaScript', '.js'),
('C++', '.hpp'), ('C++', '.cpp')]
with open('languages.txt', 'w') as file_languages:
for entry in languages:
file_languages.write(f"{entry[0]}\t{entry[1]}\n")
self.languages = read_dictionary_one_to_set('./', 'languages.txt')
def test_read_dictionary_missing_index_columns(self):
"""With two columns file, indices other than (0, 1), like (1, 2), show error."""
# Create file with three columns, some not in lexicographic order
pairs = [('a', 'b'), ('c', 'b'), ('d', 'e')]
file_name = TestCase.id(self) + '_pairs.txt'
with open(file_name, 'w') as file:
for x, y in pairs:
file.write(f"{x}\t{y}\n")
with self.assertRaises(
ValueError,
msg=
'Should raise an exception because needed columns of the file are missing.'
):
read_dictionary_one_to_set('', file_name, col_indices=(1, 2))
os.remove(file_name)
def test_read_dictionary_order_pairs_true(self):
# Create file with pairs. Some with inverted lexicographic order
pairs = [('a', 'b'), ('c', 'b'), ('d', 'e')]
file_name = TestCase.id(self) + '_pairs.txt'
with open(file_name, 'w') as file:
for x, y in pairs:
file.write(f"{x}\t{y}\n")
# Execute target method
result = read_dictionary_one_to_set('', file_name, order_pairs=True)
# Check the pairs order was corrected, showing them as key and value when word1 < word2 Lexicographical order
self.assertIn(
'b',
result.keys(),
msg="Missing key because it did not order the column values")
self.assertEqual(3, len(result.keys()), msg="Wrong number of columns")
os.remove(file_name)
def test_read_dictionary_indices_1_2(self):
# Create file with three columns, some not in lexicographic order
trios = [(1, 1, 2), (2, 3, 2), (3, 4, 5)]
file_name = TestCase.id(self) + '_pairs.txt'
with open(file_name, 'w') as file:
for x, y, z in trios:
file.write(f"{x}\t{y}\t{z}\n")
# Execute target method
result = read_dictionary_one_to_set('',
file_name,
order_pairs=True,
col_indices=(1, 2))
# Check values are correct
self.assertIn('1', result.keys(), msg="Missing key in dictionary")
self.assertIn('2', result.keys(), msg="Missing key in dictionary")
self.assertNotIn('3', result.keys(), msg="Incorrect key in dictionary")
self.assertIn('4', result.keys(), msg="Missing key in dictionary")
# Remove file
os.remove(file_name)
def create_ensembl_uniprot_mapping(config):
"""Creates a one to one dictionary"""
print("Reading Entrez -- UniProt mapping...")
temp_mapping = dictionaries.read_dictionary_one_to_set(config['PATH_STRING'], config['STRING_ID_MAP'],
order_pairs=False, col_indices=(2, 1), ignore_header=False)
return {k: {p.split('|')[0] for p in v} for k, v in temp_mapping.items()} # Extract the Uniprot accessions