def test_k_word_frequencies(self):
# overlapping = True
expected = defaultdict(int)
expected['A'] = 3 / 7.
expected['C'] = 1 / 7.
expected['G'] = 1 / 7.
expected['T'] = 2 / 7.
self.assertEqual(self.b1.k_word_frequencies(1, overlapping=True),
expected)
expected = defaultdict(int)
expected['GAT'] = 1 / 5.
expected['ATT'] = 1 / 5.
expected['TTA'] = 1 / 5.
expected['TAC'] = 1 / 5.
expected['ACA'] = 1 / 5.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=True),
expected)
# overlapping = False
expected = defaultdict(int)
expected['GAT'] = 1 / 2.
expected['TAC'] = 1 / 2.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=False),
expected)
expected = defaultdict(int)
expected['GATTACA'] = 1.0
self.assertEqual(self.b1.k_word_frequencies(7, overlapping=False),
expected)
expected = defaultdict(int)
empty = BiologicalSequence('')
self.assertEqual(empty.k_word_frequencies(1, overlapping=False),
expected)
def test_k_words(self):
# overlapping = True
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=True)),
['GA', 'AT', 'TT', 'TA', 'AC', 'CA'])
self.assertEqual(list(self.b1.k_words(3, overlapping=True)),
['GAT', 'ATT', 'TTA', 'TAC', 'ACA'])
self.assertEqual(list(self.b1.k_words(7, overlapping=True)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=True)),
[])
# overlapping = False
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=False)),
['GA', 'TT', 'AC'])
self.assertEqual(list(self.b1.k_words(3, overlapping=False)),
['GAT', 'TAC'])
self.assertEqual(list(self.b1.k_words(7, overlapping=False)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=False)),
[])
# error on invalid k
self.assertRaises(ValueError, list, self.b1.k_words(0))
self.assertRaises(ValueError, list, self.b1.k_words(-42))
# tests with different sequences
self.assertEqual(list(self.b8.k_words(3, overlapping=False)),
['HE.', '.--', '..L'])
b = BiologicalSequence('')
self.assertEqual(list(b.k_words(3)), [])
def test_k_words(self):
# overlapping = True
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=True)),
['GA', 'AT', 'TT', 'TA', 'AC', 'CA'])
self.assertEqual(list(self.b1.k_words(3, overlapping=True)),
['GAT', 'ATT', 'TTA', 'TAC', 'ACA'])
self.assertEqual(list(self.b1.k_words(7, overlapping=True)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=True)), [])
# overlapping = False
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=False)),
['GA', 'TT', 'AC'])
self.assertEqual(list(self.b1.k_words(3, overlapping=False)),
['GAT', 'TAC'])
self.assertEqual(list(self.b1.k_words(7, overlapping=False)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=False)), [])
# error on invalid k
self.assertRaises(ValueError, list, self.b1.k_words(0))
self.assertRaises(ValueError, list, self.b1.k_words(-42))
# tests with different sequences
self.assertEqual(list(self.b8.k_words(3, overlapping=False)),
['HE.', '.--', '..L'])
b = BiologicalSequence('')
self.assertEqual(list(b.k_words(3)), [])
def test_k_word_frequencies(self):
# overlapping = True
expected = defaultdict(int)
expected['A'] = 3/7.
expected['C'] = 1/7.
expected['G'] = 1/7.
expected['T'] = 2/7.
self.assertEqual(self.b1.k_word_frequencies(1, overlapping=True),
expected)
expected = defaultdict(int)
expected['GAT'] = 1/5.
expected['ATT'] = 1/5.
expected['TTA'] = 1/5.
expected['TAC'] = 1/5.
expected['ACA'] = 1/5.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=True),
expected)
# overlapping = False
expected = defaultdict(int)
expected['GAT'] = 1/2.
expected['TAC'] = 1/2.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=False),
expected)
expected = defaultdict(int)
expected['GATTACA'] = 1.0
self.assertEqual(self.b1.k_word_frequencies(7, overlapping=False),
expected)
expected = defaultdict(int)
empty = BiologicalSequence('')
self.assertEqual(empty.k_word_frequencies(1, overlapping=False),
expected)
def setUp(self):
"""Setup for Fasta tests."""
self.strings = ['AAAA', 'CCCC', 'gggg', 'uuuu']
self.fasta_no_label = '>0\nAAAA\n>1\nCCCC\n>2\ngggg\n>3\nuuuu'
self.fasta_with_label =\
'>1st\nAAAA\n>2nd\nCCCC\n>3rd\nGGGG\n>4th\nUUUU'
self.fasta_with_label_lw2 =\
'>1st\nAA\nAA\n>2nd\nCC\nCC\n>3rd\nGG\nGG\n>4th\nUU\nUU'
self.alignment_dict = {
'1st': 'AAAA',
'2nd': 'CCCC',
'3rd': 'GGGG',
'4th': 'UUUU'
}
self.sequence_objects_a = [
DNASequence('ACTCGAGATC', 'seq1'),
DNASequence('GGCCT', 'seq2')
]
self.sequence_objects_b = [
BiologicalSequence('ACTCGAGATC', 'seq1'),
BiologicalSequence('GGCCT', 'seq2')
]
seqs = [
DNASequence("ACC--G-GGTA..", id="seq1"),
DNASequence("TCC--G-GGCA..", id="seqs2")
]
self.alignment = Alignment(seqs)
def test_init_varied_input(self):
# init as string
b = BiologicalSequence('ACCGGXZY')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
# init as string with optional values
b = BiologicalSequence('ACCGGXZY', 'test-seq-1',
'The first test sequence')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "test-seq-1")
self.assertEqual(b.description, "The first test sequence")
# test init as a different string
b = BiologicalSequence('WRRTY')
self.assertEqual(str(b), 'WRRTY')
# init as list
b = BiologicalSequence(list('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
# init as tuple
b = BiologicalSequence(tuple('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
def test_eq(self):
self.assertTrue(self.b1 == self.b1)
self.assertTrue(self.b2 == self.b2)
self.assertTrue(self.b3 == self.b3)
self.assertTrue(self.b1 != self.b3)
self.assertTrue(self.b1 != self.b2)
self.assertTrue(self.b2 != self.b3)
# identicial sequences of the same type are equal, even if they have
# different ids and/or descriptions
self.assertTrue(
BiologicalSequence('ACGT') == BiologicalSequence('ACGT'))
self.assertTrue(
BiologicalSequence('ACGT', id='a') == BiologicalSequence('ACGT',
id='b'))
self.assertTrue(
BiologicalSequence('ACGT', description='c') == BiologicalSequence(
'ACGT', description='d'))
self.assertTrue(
BiologicalSequence('ACGT', id='a', description='c') ==
BiologicalSequence('ACGT', id='b', description='d'))
# different type causes sequences to not be equal
self.assertFalse(
BiologicalSequence('ACGT') == NucleotideSequence('ACGT'))
def get_representatives(mapping, seqs):
"""Returns representative seqs.
mapping: The prefix mapping dict
seqs_fh: An open Fasta filehandle
"""
for (label, seq) in seqs:
if(label in mapping):
seq = BiologicalSequence(
seq, id="%s: %d" % (label, len(mapping[label]) + 1))
yield seq.upper()
def get_representatives(mapping, seqs):
"""Returns representative seqs.
mapping: The prefix mapping dict
seqs_fh: An open Fasta filehandle
"""
for (label, seq) in seqs:
if (label in mapping):
seq = BiologicalSequence(seq,
id="%s: %d" %
(label, len(mapping[label]) + 1))
yield seq.upper()
def test_init_with_validation(self):
self.assertRaises(BiologicalSequenceError,
BiologicalSequence,
"ACC",
validate=True)
# no error raised when only allow characters are passed
BiologicalSequence("..--..", validate=True)
def test_distance(self):
# note that test_hamming_distance covers default behavior more
# extensively
self.assertEqual(self.b1.distance(self.b1), 0.0)
self.assertEqual(self.b1.distance(BiologicalSequence('GATTACC')), 1./7)
def dumb_distance(x, y):
return 42
self.assertEqual(
self.b1.distance(self.b1, distance_fn=dumb_distance), 42)
def write_Fasta_from_name_seq_pairs(name_seqs, fh):
"""writes a list of (name,seqs) to filehandle.
name_seqs: (name,seqs) pair such as from parse_fasta
fh: an open filehandle
"""
if fh is None:
raise ValueError("Need open file handle to write to.")
for (name, seq) in name_seqs:
fh.write("%s\n" % BiologicalSequence(seq, identifier=name).to_fasta())
def combine_mappings(fasta_fh, mapping_fh, denoised_seqs_fh,
otu_picker_otu_map_fh, out_dir):
"""Combine denoiser and OTU picker mapping file, replace flowgram IDs.
fasta_fh: a fasta file with labels as produced by Qiime's split_libraries.py
used to replace flowgram id with the unique se_sample_id
mapping_fh: The cluster mapping from the denoiser.py
denoised_seqs_fh: the Fasta output files from denoiser.py
otu_picker_map_fh: cluster map from otu picker on denoised_seqs_fh
out_dir: output directory
"""
# read in mapping from split_library file
labels = imap(lambda a_b: a_b[0], parse_fasta(fasta_fh))
# mapping from seq_id to sample_id
sample_id_mapping = extract_read_to_sample_mapping(labels)
denoiser_mapping = read_denoiser_mapping(mapping_fh)
# read in cd_hit otu map
# and write out combined otu_picker+denoiser map
otu_fh = open(out_dir + "/denoised_otu_map.txt", "w")
for otu_line in otu_picker_otu_map_fh:
otu_split = otu_line.split()
otu = otu_split[0]
ids = otu_split[1:]
get_sample_id = sample_id_mapping.get
# concat lists
# make sure the biggest one is first for pick_repr
all_ids = sort_ids(ids, denoiser_mapping)
all_ids.extend(sum([denoiser_mapping[id] for id in ids], []))
try:
otu_fh.write("%s\t" % otu +
"\t".join(map(get_sample_id, all_ids)) + "\n")
except TypeError:
# get returns Null if denoiser_mapping id not present in
# sample_id_mapping
print "Found id in denoiser output, which was not found in split_libraries " +\
"output FASTA file. Wrong file?"
exit()
fasta_out_fh = open(out_dir + "/denoised_all.fasta", "w")
for label, seq in parse_fasta(denoised_seqs_fh):
id = label.split()[0]
newlabel = "%s %s" % (sample_id_mapping[id], id)
fasta_out_fh.write(BiologicalSequence(seq, id=newlabel).to_fasta())
def setUp(self):
self.b1 = BiologicalSequence('GATTACA')
self.b2 = BiologicalSequence(
'ACCGGTACC', id="test-seq-2",
description="A test sequence")
self.b3 = BiologicalSequence(
'GREG', id="test-seq-3", description="A protein sequence")
self.b4 = BiologicalSequence(
'PRTEIN', id="test-seq-4")
self.b5 = BiologicalSequence(
'LLPRTEIN', description="some description")
self.b6 = BiologicalSequence('ACGTACGTACGT')
self.b7 = BiologicalSequence('..--..')
self.b8 = BiologicalSequence('HE..--..LLO')
def setUp(self):
""" Initialize values to be used in tests
"""
self.b1 = BiologicalSequence('GATTACA')
self.b2 = BiologicalSequence(
'ACCGGTACC', identifier="test-seq-2",
description="A test sequence")
self.b3 = BiologicalSequence(
'GREG', identifier="test-seq-3", description="A protein sequence")
self.b4 = BiologicalSequence(
'PRTEIN', identifier="test-seq-4")
self.b5 = BiologicalSequence(
'LLPRTEIN', description="some description")
self.b6 = BiologicalSequence('ACGTACGTACGT')
self.b7 = BiologicalSequence('..--..')
self.b8 = BiologicalSequence('HE..--..LLO')
def test_gap_maps(self):
# in sequence with no gaps, the gap_maps are identical
self.assertEqual(self.b1.gap_maps(),
([0, 1, 2, 3, 4, 5, 6], [0, 1, 2, 3, 4, 5, 6]))
# in sequence with all gaps, the map of degapped to gapped is the empty
# list (bc its length is 0), and the map of gapped to degapped is all
# None
self.assertEqual(self.b7.gap_maps(),
([], [None, None, None, None, None, None]))
self.assertEqual(self.b8.gap_maps(),
([0, 1, 8, 9, 10],
[0, 1, None, None, None, None, None, None, 2, 3, 4]))
# example from the gap_maps doc string
self.assertEqual(
BiologicalSequence('-ACCGA-TA-').gap_maps(),
([1, 2, 3, 4, 5, 7, 8], [None, 0, 1, 2, 3, 4, None, 5, 6, None]))
class BiologicalSequenceTests(TestCase):
""" Tests of the BiologicalSequence class """
def setUp(self):
""" Initialize values to be used in tests
"""
self.b1 = BiologicalSequence('GATTACA')
self.b2 = BiologicalSequence(
'ACCGGTACC', identifier="test-seq-2",
description="A test sequence")
self.b3 = BiologicalSequence(
'GREG', identifier="test-seq-3", description="A protein sequence")
self.b4 = BiologicalSequence(
'PRTEIN', identifier="test-seq-4")
self.b5 = BiologicalSequence(
'LLPRTEIN', description="some description")
self.b6 = BiologicalSequence('ACGTACGTACGT')
self.b7 = BiologicalSequence('..--..')
self.b8 = BiologicalSequence('HE..--..LLO')
def test_init(self):
""" Initialization functions as expected with varied input types
"""
# init as string
b = BiologicalSequence('ACCGGXZY')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.identifier, "")
self.assertEqual(b.description, "")
# init as string with optional values
b = BiologicalSequence(
'ACCGGXZY', 'test-seq-1', 'The first test sequence')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.identifier, "test-seq-1")
self.assertEqual(b.description, "The first test sequence")
# test init as a different string
b = BiologicalSequence('WRRTY')
self.assertEqual(str(b), 'WRRTY')
# init as list
b = BiologicalSequence(list('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.identifier, "")
self.assertEqual(b.description, "")
# init as tuple
b = BiologicalSequence(tuple('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.identifier, "")
self.assertEqual(b.description, "")
def test_init_validate(self):
""" initialization with validation functions as expected
"""
self.assertRaises(BiologicalSequenceError, BiologicalSequence, "ACC",
validate=True)
# no error raised when only allow characters are passed
BiologicalSequence("..--..", validate=True)
def test_contains(self):
""" contains functions as expected
"""
self.assertTrue('G' in self.b1)
self.assertFalse('g' in self.b1)
def test_eq(self):
""" equality functions as expected
"""
self.assertTrue(self.b1 == self.b1)
self.assertTrue(self.b2 == self.b2)
self.assertTrue(self.b3 == self.b3)
self.assertTrue(self.b1 != self.b3)
self.assertTrue(self.b1 != self.b2)
self.assertTrue(self.b2 != self.b3)
# identicial sequences of the same type are equal, even if they have
# different identifiers and/or descriptions
self.assertTrue(
BiologicalSequence('ACGT') == BiologicalSequence('ACGT'))
self.assertTrue(
BiologicalSequence('ACGT', identifier='a') ==
BiologicalSequence('ACGT', identifier='b'))
self.assertTrue(
BiologicalSequence('ACGT', description='c') ==
BiologicalSequence('ACGT', description='d'))
self.assertTrue(
BiologicalSequence('ACGT', identifier='a', description='c') ==
BiologicalSequence('ACGT', identifier='b', description='d'))
# different type causes sequences to not be equal
self.assertFalse(
BiologicalSequence('ACGT') == NucleotideSequence('ACGT'))
def test_getitem(self):
""" getitem functions as expected
"""
self.assertEqual(self.b1[0], BiologicalSequence('G'))
self.assertEqual(self.b1[:], BiologicalSequence('GATTACA'))
self.assertEqual(self.b1[::-1], BiologicalSequence('ACATTAG'))
def test_iter(self):
""" iter functions as expected
"""
b1_iter = iter(self.b1)
for actual, expected in zip(b1_iter, "GATTACA"):
self.assertEqual(actual, expected)
self.assertRaises(StopIteration, b1_iter.next)
def test_len(self):
""" len functions as expected
"""
self.assertEqual(len(self.b1), 7)
self.assertEqual(len(self.b2), 9)
self.assertEqual(len(self.b3), 4)
def test_repr(self):
""" repr functions as expected
"""
self.assertEqual(repr(self.b1),
"<BiologicalSequence: GATTACA (length: 7)>")
self.assertEqual(repr(self.b6),
"<BiologicalSequence: ACGTACGTAC... (length: 12)>")
def test_reversed(self):
""" reversed functions as expected
"""
b1_reversed = reversed(self.b1)
for actual, expected in zip(b1_reversed, "ACATTAG"):
self.assertEqual(actual, expected)
self.assertRaises(StopIteration, b1_reversed.next)
def test_str(self):
""" str functions as expected
"""
self.assertEqual(str(self.b1), "GATTACA")
self.assertEqual(str(self.b2), "ACCGGTACC")
self.assertEqual(str(self.b3), "GREG")
def test_alphabet(self):
""" alphabet property functions as expected
"""
self.assertEqual(self.b1.alphabet(), set())
def test_description(self):
""" description property functions as expected
"""
self.assertEqual(self.b1.description, "")
self.assertEqual(self.b2.description, "A test sequence")
self.assertEqual(self.b3.description, "A protein sequence")
def test_gap_alphabet(self):
""" gap_alphabet property functions as expected
"""
self.assertEqual(self.b1.gap_alphabet(), set('-.'))
def test_identifier(self):
""" identifier property functions as expected
"""
self.assertEqual(self.b1.identifier, "")
self.assertEqual(self.b2.identifier, "test-seq-2")
self.assertEqual(self.b3.identifier, "test-seq-3")
def test_count(self):
""" count functions as expected
"""
self.assertEqual(self.b1.count('A'), 3)
self.assertEqual(self.b1.count('T'), 2)
self.assertEqual(self.b1.count('TT'), 1)
def test_degap(self):
""" degap functions as expected
"""
self.assertEqual(self.b1.degap(), self.b1)
self.assertEqual(self.b7.degap(), BiologicalSequence(''))
self.assertEqual(self.b8.degap(), BiologicalSequence('HELLO'))
def test_distance(self):
""" distance functions as expected
"""
# note that test_hamming_distance covers default behavior more
# extensively
self.assertEqual(self.b1.distance(self.b1), 0.0)
self.assertEqual(self.b1.distance(BiologicalSequence('GATTACC')), 1./7)
def dumb_distance(x, y):
return 42
self.assertEqual(
self.b1.distance(self.b1, distance_fn=dumb_distance), 42)
def test_fraction_diff(self):
""" fraction_diff functions as expected
"""
self.assertEqual(self.b1.fraction_diff(self.b1), 0., 5)
self.assertEqual(
self.b1.fraction_diff(BiologicalSequence('GATTACC')), 1. / 7., 5)
def test_fraction_same(self):
""" fraction_same functions as expected
"""
self.assertAlmostEqual(self.b1.fraction_same(self.b1), 1., 5)
self.assertAlmostEqual(
self.b1.fraction_same(BiologicalSequence('GATTACC')), 6. / 7., 5)
def test_gap_maps(self):
""" gap_maps functions as expected
"""
# in sequence with no gaps, the gap_maps are identical
self.assertEqual(self.b1.gap_maps(),
([0, 1, 2, 3, 4, 5, 6], [0, 1, 2, 3, 4, 5, 6]))
# in sequence with all gaps, the map of degapped to gapped is the empty
# list (bc its length is 0), and the map of gapped to degapped is all
# None
self.assertEqual(self.b7.gap_maps(),
([], [None, None, None, None, None, None]))
self.assertEqual(self.b8.gap_maps(),
([0, 1, 8, 9, 10],
[0, 1, None, None, None, None, None, None, 2, 3, 4]))
# example from the gap_maps doc string
self.assertEqual(BiologicalSequence('-ACCGA-TA-').gap_maps(),
([1, 2, 3, 4, 5, 7, 8],
[None, 0, 1, 2, 3, 4, None, 5, 6, None]))
def test_gap_vector(self):
""" gap_vector functions as expected
"""
self.assertEqual(self.b1.gap_vector(),
[False] * len(self.b1))
self.assertEqual(self.b7.gap_vector(),
[True] * len(self.b7))
self.assertEqual(self.b8.gap_vector(),
[False, False, True, True, True, True,
True, True, False, False, False])
def test_unsupported_characters(self):
""" unsupported_characters functions as expected
"""
self.assertEqual(self.b1.unsupported_characters(), set('GATC'))
self.assertEqual(self.b7.unsupported_characters(), set())
def test_has_unsupported_characters(self):
""" has_unsupported_characters functions as expected
"""
self.assertTrue(self.b1.has_unsupported_characters())
self.assertFalse(self.b7.has_unsupported_characters())
def test_index(self):
""" index functions as expected """
self.assertEqual(self.b1.index('G'), 0)
self.assertEqual(self.b1.index('A'), 1)
self.assertEqual(self.b1.index('AC'), 4)
self.assertRaises(ValueError, self.b1.index, 'x')
def test_is_gap(self):
""" is_gap functions as expected """
self.assertTrue(self.b1.is_gap('.'))
self.assertTrue(self.b1.is_gap('-'))
self.assertFalse(self.b1.is_gap('A'))
self.assertFalse(self.b1.is_gap('x'))
self.assertFalse(self.b1.is_gap(' '))
self.assertFalse(self.b1.is_gap(''))
def test_is_gapped(self):
""" is_gapped functions as expected """
self.assertFalse(self.b1.is_gapped())
self.assertFalse(self.b2.is_gapped())
self.assertTrue(self.b7.is_gapped())
self.assertTrue(self.b8.is_gapped())
def test_is_valid(self):
""" is_valid functions as expected
"""
self.assertFalse(self.b1.is_valid())
self.assertTrue(self.b7.is_valid())
def test_to_fasta(self):
""" to_fasta functions as expected
"""
self.assertEqual(self.b1.to_fasta(), ">\nGATTACA\n")
self.assertEqual(self.b1.to_fasta(terminal_character=""), ">\nGATTACA")
self.assertEqual(self.b2.to_fasta(),
">test-seq-2 A test sequence\nACCGGTACC\n")
self.assertEqual(self.b3.to_fasta(),
">test-seq-3 A protein sequence\nGREG\n")
self.assertEqual(self.b4.to_fasta(),
">test-seq-4\nPRTEIN\n")
self.assertEqual(self.b5.to_fasta(),
"> some description\nLLPRTEIN\n")
# alt parameters
self.assertEqual(self.b2.to_fasta(field_delimiter=":"),
">test-seq-2:A test sequence\nACCGGTACC\n")
self.assertEqual(self.b2.to_fasta(terminal_character="!"),
">test-seq-2 A test sequence\nACCGGTACC!")
self.assertEqual(
self.b2.to_fasta(field_delimiter=":", terminal_character="!"),
">test-seq-2:A test sequence\nACCGGTACC!")
def test_upper(self):
""" upper functions as expected
"""
b = NucleotideSequence('GAt.ACa-', identifier='x', description='42')
expected = NucleotideSequence('GAT.ACA-', identifier='x',
description='42')
self.assertEqual(b.upper(), expected)
def test_lower(self):
""" lower functions as expected
"""
b = NucleotideSequence('GAt.ACa-', identifier='x', description='42')
expected = NucleotideSequence('gat.aca-', identifier='x',
description='42')
self.assertEqual(b.lower(), expected)
def hamming_distance(s1, s2):
s1 = BiologicalSequence(s1)
s2 = BiologicalSequence(s2)
return s1.distance(s2)
class BiologicalSequenceTests(TestCase):
def setUp(self):
self.b1 = BiologicalSequence('GATTACA')
self.b2 = BiologicalSequence(
'ACCGGTACC', id="test-seq-2",
description="A test sequence")
self.b3 = BiologicalSequence(
'GREG', id="test-seq-3", description="A protein sequence")
self.b4 = BiologicalSequence(
'PRTEIN', id="test-seq-4")
self.b5 = BiologicalSequence(
'LLPRTEIN', description="some description")
self.b6 = BiologicalSequence('ACGTACGTACGT')
self.b7 = BiologicalSequence('..--..')
self.b8 = BiologicalSequence('HE..--..LLO')
def test_init_varied_input(self):
# init as string
b = BiologicalSequence('ACCGGXZY')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
# init as string with optional values
b = BiologicalSequence(
'ACCGGXZY', 'test-seq-1', 'The first test sequence')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "test-seq-1")
self.assertEqual(b.description, "The first test sequence")
# test init as a different string
b = BiologicalSequence('WRRTY')
self.assertEqual(str(b), 'WRRTY')
# init as list
b = BiologicalSequence(list('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
# init as tuple
b = BiologicalSequence(tuple('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
def test_init_with_validation(self):
self.assertRaises(BiologicalSequenceError, BiologicalSequence, "ACC",
validate=True)
try:
# no error raised when only allow characters are passed
BiologicalSequence("..--..", validate=True)
except BiologicalSequenceError:
self.assertTrue(False)
def test_contains(self):
self.assertTrue('G' in self.b1)
self.assertFalse('g' in self.b1)
def test_eq(self):
self.assertTrue(self.b1 == self.b1)
self.assertTrue(self.b2 == self.b2)
self.assertTrue(self.b3 == self.b3)
self.assertTrue(self.b1 != self.b3)
self.assertTrue(self.b1 != self.b2)
self.assertTrue(self.b2 != self.b3)
# identicial sequences of the same type are equal, even if they have
# different ids and/or descriptions
self.assertTrue(
BiologicalSequence('ACGT') == BiologicalSequence('ACGT'))
self.assertTrue(
BiologicalSequence('ACGT', id='a') ==
BiologicalSequence('ACGT', id='b'))
self.assertTrue(
BiologicalSequence('ACGT', description='c') ==
BiologicalSequence('ACGT', description='d'))
self.assertTrue(
BiologicalSequence('ACGT', id='a', description='c') ==
BiologicalSequence('ACGT', id='b', description='d'))
# different type causes sequences to not be equal
self.assertFalse(
BiologicalSequence('ACGT') == NucleotideSequence('ACGT'))
def test_getitem(self):
self.assertEqual(self.b1[0], BiologicalSequence('G'))
self.assertEqual(self.b1[:], BiologicalSequence('GATTACA'))
self.assertEqual(self.b1[::-1], BiologicalSequence('ACATTAG'))
def test_getitem_out_of_range(self):
with self.assertRaises(IndexError):
self.b1[42]
def test_hash(self):
self.assertTrue(isinstance(hash(self.b1), int))
def test_iter(self):
b1_iter = iter(self.b1)
for actual, expected in zip(b1_iter, "GATTACA"):
self.assertEqual(actual, expected)
self.assertRaises(StopIteration, lambda: next(b1_iter))
def test_k_words(self):
# overlapping = True
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=True)),
['GA', 'AT', 'TT', 'TA', 'AC', 'CA'])
self.assertEqual(list(self.b1.k_words(3, overlapping=True)),
['GAT', 'ATT', 'TTA', 'TAC', 'ACA'])
self.assertEqual(list(self.b1.k_words(7, overlapping=True)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=True)),
[])
# overlapping = False
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=False)),
['GA', 'TT', 'AC'])
self.assertEqual(list(self.b1.k_words(3, overlapping=False)),
['GAT', 'TAC'])
self.assertEqual(list(self.b1.k_words(7, overlapping=False)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=False)),
[])
# error on invalid k
self.assertRaises(ValueError, list, self.b1.k_words(0))
self.assertRaises(ValueError, list, self.b1.k_words(-42))
# tests with different sequences
self.assertEqual(list(self.b8.k_words(3, overlapping=False)),
['HE.', '.--', '..L'])
b = BiologicalSequence('')
self.assertEqual(list(b.k_words(3)), [])
def test_k_word_counts(self):
# overlapping = True
expected = Counter('GATTACA')
self.assertEqual(self.b1.k_word_counts(1, overlapping=True),
expected)
expected = Counter(['GAT', 'ATT', 'TTA', 'TAC', 'ACA'])
self.assertEqual(self.b1.k_word_counts(3, overlapping=True),
expected)
# overlapping = False
expected = Counter(['GAT', 'TAC'])
self.assertEqual(self.b1.k_word_counts(3, overlapping=False),
expected)
expected = Counter(['GATTACA'])
self.assertEqual(self.b1.k_word_counts(7, overlapping=False),
expected)
def test_k_word_frequencies(self):
# overlapping = True
expected = defaultdict(int)
expected['A'] = 3/7.
expected['C'] = 1/7.
expected['G'] = 1/7.
expected['T'] = 2/7.
self.assertEqual(self.b1.k_word_frequencies(1, overlapping=True),
expected)
expected = defaultdict(int)
expected['GAT'] = 1/5.
expected['ATT'] = 1/5.
expected['TTA'] = 1/5.
expected['TAC'] = 1/5.
expected['ACA'] = 1/5.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=True),
expected)
# overlapping = False
expected = defaultdict(int)
expected['GAT'] = 1/2.
expected['TAC'] = 1/2.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=False),
expected)
expected = defaultdict(int)
expected['GATTACA'] = 1.0
self.assertEqual(self.b1.k_word_frequencies(7, overlapping=False),
expected)
expected = defaultdict(int)
empty = BiologicalSequence('')
self.assertEqual(empty.k_word_frequencies(1, overlapping=False),
expected)
def test_len(self):
self.assertEqual(len(self.b1), 7)
self.assertEqual(len(self.b2), 9)
self.assertEqual(len(self.b3), 4)
def test_repr(self):
self.assertEqual(repr(self.b1),
"<BiologicalSequence: GATTACA (length: 7)>")
self.assertEqual(repr(self.b6),
"<BiologicalSequence: ACGTACGTAC... (length: 12)>")
def test_reversed(self):
b1_reversed = reversed(self.b1)
for actual, expected in zip(b1_reversed, "ACATTAG"):
self.assertEqual(actual, expected)
self.assertRaises(StopIteration, lambda: next(b1_reversed))
def test_str(self):
self.assertEqual(str(self.b1), "GATTACA")
self.assertEqual(str(self.b2), "ACCGGTACC")
self.assertEqual(str(self.b3), "GREG")
def test_alphabet(self):
self.assertEqual(self.b1.alphabet(), set())
def test_description(self):
self.assertEqual(self.b1.description, "")
self.assertEqual(self.b2.description, "A test sequence")
self.assertEqual(self.b3.description, "A protein sequence")
def test_gap_alphabet(self):
self.assertEqual(self.b1.gap_alphabet(), set('-.'))
def test_id(self):
self.assertEqual(self.b1.id, "")
self.assertEqual(self.b2.id, "test-seq-2")
self.assertEqual(self.b3.id, "test-seq-3")
def test_count(self):
self.assertEqual(self.b1.count('A'), 3)
self.assertEqual(self.b1.count('T'), 2)
self.assertEqual(self.b1.count('TT'), 1)
def test_degap(self):
self.assertEqual(self.b1.degap(), self.b1)
self.assertEqual(self.b7.degap(), BiologicalSequence(''))
self.assertEqual(self.b8.degap(), BiologicalSequence('HELLO'))
def test_distance(self):
# note that test_hamming_distance covers default behavior more
# extensively
self.assertEqual(self.b1.distance(self.b1), 0.0)
self.assertEqual(self.b1.distance(BiologicalSequence('GATTACC')), 1./7)
def dumb_distance(x, y):
return 42
self.assertEqual(
self.b1.distance(self.b1, distance_fn=dumb_distance), 42)
def test_distance_unequal_length(self):
# Hamming distance (default) requires that sequences are of equal
# length
with self.assertRaises(BiologicalSequenceError):
self.b1.distance(self.b2)
# alternate distance functions don't have that requirement (unless
# it's implemented within the provided distance function)
def dumb_distance(x, y):
return 42
self.assertEqual(
self.b1.distance(self.b2, distance_fn=dumb_distance), 42)
def test_fraction_diff(self):
self.assertEqual(self.b1.fraction_diff(self.b1), 0., 5)
self.assertEqual(
self.b1.fraction_diff(BiologicalSequence('GATTACC')), 1. / 7., 5)
def test_fraction_same(self):
self.assertAlmostEqual(self.b1.fraction_same(self.b1), 1., 5)
self.assertAlmostEqual(
self.b1.fraction_same(BiologicalSequence('GATTACC')), 6. / 7., 5)
def test_gap_maps(self):
# in sequence with no gaps, the gap_maps are identical
self.assertEqual(self.b1.gap_maps(),
([0, 1, 2, 3, 4, 5, 6], [0, 1, 2, 3, 4, 5, 6]))
# in sequence with all gaps, the map of degapped to gapped is the empty
# list (bc its length is 0), and the map of gapped to degapped is all
# None
self.assertEqual(self.b7.gap_maps(),
([], [None, None, None, None, None, None]))
self.assertEqual(self.b8.gap_maps(),
([0, 1, 8, 9, 10],
[0, 1, None, None, None, None, None, None, 2, 3, 4]))
# example from the gap_maps doc string
self.assertEqual(BiologicalSequence('-ACCGA-TA-').gap_maps(),
([1, 2, 3, 4, 5, 7, 8],
[None, 0, 1, 2, 3, 4, None, 5, 6, None]))
def test_gap_vector(self):
self.assertEqual(self.b1.gap_vector(),
[False] * len(self.b1))
self.assertEqual(self.b7.gap_vector(),
[True] * len(self.b7))
self.assertEqual(self.b8.gap_vector(),
[False, False, True, True, True, True,
True, True, False, False, False])
def test_unsupported_characters(self):
self.assertEqual(self.b1.unsupported_characters(), set('GATC'))
self.assertEqual(self.b7.unsupported_characters(), set())
def test_has_unsupported_characters(self):
self.assertTrue(self.b1.has_unsupported_characters())
self.assertFalse(self.b7.has_unsupported_characters())
def test_index(self):
""" index functions as expected """
self.assertEqual(self.b1.index('G'), 0)
self.assertEqual(self.b1.index('A'), 1)
self.assertEqual(self.b1.index('AC'), 4)
self.assertRaises(ValueError, self.b1.index, 'x')
def test_is_gap(self):
self.assertTrue(self.b1.is_gap('.'))
self.assertTrue(self.b1.is_gap('-'))
self.assertFalse(self.b1.is_gap('A'))
self.assertFalse(self.b1.is_gap('x'))
self.assertFalse(self.b1.is_gap(' '))
self.assertFalse(self.b1.is_gap(''))
def test_is_gapped(self):
self.assertFalse(self.b1.is_gapped())
self.assertFalse(self.b2.is_gapped())
self.assertTrue(self.b7.is_gapped())
self.assertTrue(self.b8.is_gapped())
def test_is_valid(self):
self.assertFalse(self.b1.is_valid())
self.assertTrue(self.b7.is_valid())
def test_to_fasta(self):
self.assertEqual(self.b1.to_fasta(), ">\nGATTACA\n")
self.assertEqual(self.b1.to_fasta(terminal_character=""), ">\nGATTACA")
self.assertEqual(self.b2.to_fasta(),
">test-seq-2 A test sequence\nACCGGTACC\n")
self.assertEqual(self.b3.to_fasta(),
">test-seq-3 A protein sequence\nGREG\n")
self.assertEqual(self.b4.to_fasta(),
">test-seq-4\nPRTEIN\n")
self.assertEqual(self.b5.to_fasta(),
"> some description\nLLPRTEIN\n")
# alt parameters
self.assertEqual(self.b2.to_fasta(field_delimiter=":"),
">test-seq-2:A test sequence\nACCGGTACC\n")
self.assertEqual(self.b2.to_fasta(terminal_character="!"),
">test-seq-2 A test sequence\nACCGGTACC!")
self.assertEqual(
self.b2.to_fasta(field_delimiter=":", terminal_character="!"),
">test-seq-2:A test sequence\nACCGGTACC!")
def test_upper(self):
b = NucleotideSequence('GAt.ACa-', id='x', description='42')
expected = NucleotideSequence('GAT.ACA-', id='x',
description='42')
self.assertEqual(b.upper(), expected)
def test_lower(self):
b = NucleotideSequence('GAt.ACa-', id='x', description='42')
expected = NucleotideSequence('gat.aca-', id='x',
description='42')
self.assertEqual(b.lower(), expected)
def hamming_distance(s1, s2):
s1 = BiologicalSequence(s1)
s2 = BiologicalSequence(s2)
return s1.distance(s2)
def test_getitem(self):
self.assertEqual(self.b1[0], BiologicalSequence('G'))
self.assertEqual(self.b1[:], BiologicalSequence('GATTACA'))
self.assertEqual(self.b1[::-1], BiologicalSequence('ACATTAG'))
class BiologicalSequenceTests(TestCase):
def setUp(self):
self.b1 = BiologicalSequence('GATTACA')
self.b2 = BiologicalSequence('ACCGGTACC',
id="test-seq-2",
description="A test sequence")
self.b3 = BiologicalSequence('GREG',
id="test-seq-3",
description="A protein sequence")
self.b4 = BiologicalSequence('PRTEIN', id="test-seq-4")
self.b5 = BiologicalSequence('LLPRTEIN',
description="some description")
self.b6 = BiologicalSequence('ACGTACGTACGT')
self.b7 = BiologicalSequence('..--..')
self.b8 = BiologicalSequence('HE..--..LLO')
def test_init_varied_input(self):
# init as string
b = BiologicalSequence('ACCGGXZY')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
# init as string with optional values
b = BiologicalSequence('ACCGGXZY', 'test-seq-1',
'The first test sequence')
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "test-seq-1")
self.assertEqual(b.description, "The first test sequence")
# test init as a different string
b = BiologicalSequence('WRRTY')
self.assertEqual(str(b), 'WRRTY')
# init as list
b = BiologicalSequence(list('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
# init as tuple
b = BiologicalSequence(tuple('ACCGGXZY'))
self.assertEqual(str(b), 'ACCGGXZY')
self.assertEqual(b.id, "")
self.assertEqual(b.description, "")
def test_init_with_validation(self):
self.assertRaises(BiologicalSequenceError,
BiologicalSequence,
"ACC",
validate=True)
# no error raised when only allow characters are passed
BiologicalSequence("..--..", validate=True)
def test_contains(self):
self.assertTrue('G' in self.b1)
self.assertFalse('g' in self.b1)
def test_eq(self):
self.assertTrue(self.b1 == self.b1)
self.assertTrue(self.b2 == self.b2)
self.assertTrue(self.b3 == self.b3)
self.assertTrue(self.b1 != self.b3)
self.assertTrue(self.b1 != self.b2)
self.assertTrue(self.b2 != self.b3)
# identicial sequences of the same type are equal, even if they have
# different ids and/or descriptions
self.assertTrue(
BiologicalSequence('ACGT') == BiologicalSequence('ACGT'))
self.assertTrue(
BiologicalSequence('ACGT', id='a') == BiologicalSequence('ACGT',
id='b'))
self.assertTrue(
BiologicalSequence('ACGT', description='c') == BiologicalSequence(
'ACGT', description='d'))
self.assertTrue(
BiologicalSequence('ACGT', id='a', description='c') ==
BiologicalSequence('ACGT', id='b', description='d'))
# different type causes sequences to not be equal
self.assertFalse(
BiologicalSequence('ACGT') == NucleotideSequence('ACGT'))
def test_getitem(self):
self.assertEqual(self.b1[0], BiologicalSequence('G'))
self.assertEqual(self.b1[:], BiologicalSequence('GATTACA'))
self.assertEqual(self.b1[::-1], BiologicalSequence('ACATTAG'))
def test_iter(self):
b1_iter = iter(self.b1)
for actual, expected in zip(b1_iter, "GATTACA"):
self.assertEqual(actual, expected)
self.assertRaises(StopIteration, lambda: next(b1_iter))
def test_k_words(self):
# overlapping = True
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=True)),
['GA', 'AT', 'TT', 'TA', 'AC', 'CA'])
self.assertEqual(list(self.b1.k_words(3, overlapping=True)),
['GAT', 'ATT', 'TTA', 'TAC', 'ACA'])
self.assertEqual(list(self.b1.k_words(7, overlapping=True)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=True)), [])
# overlapping = False
self.assertEqual(list(self.b1.k_words(1, overlapping=True)),
['G', 'A', 'T', 'T', 'A', 'C', 'A'])
self.assertEqual(list(self.b1.k_words(2, overlapping=False)),
['GA', 'TT', 'AC'])
self.assertEqual(list(self.b1.k_words(3, overlapping=False)),
['GAT', 'TAC'])
self.assertEqual(list(self.b1.k_words(7, overlapping=False)),
['GATTACA'])
self.assertEqual(list(self.b1.k_words(8, overlapping=False)), [])
# error on invalid k
self.assertRaises(ValueError, list, self.b1.k_words(0))
self.assertRaises(ValueError, list, self.b1.k_words(-42))
# tests with different sequences
self.assertEqual(list(self.b8.k_words(3, overlapping=False)),
['HE.', '.--', '..L'])
b = BiologicalSequence('')
self.assertEqual(list(b.k_words(3)), [])
def test_k_word_counts(self):
# overlapping = True
expected = Counter('GATTACA')
self.assertEqual(self.b1.k_word_counts(1, overlapping=True), expected)
expected = Counter(['GAT', 'ATT', 'TTA', 'TAC', 'ACA'])
self.assertEqual(self.b1.k_word_counts(3, overlapping=True), expected)
# overlapping = False
expected = Counter(['GAT', 'TAC'])
self.assertEqual(self.b1.k_word_counts(3, overlapping=False), expected)
expected = Counter(['GATTACA'])
self.assertEqual(self.b1.k_word_counts(7, overlapping=False), expected)
def test_k_word_frequencies(self):
# overlapping = True
expected = defaultdict(int)
expected['A'] = 3 / 7.
expected['C'] = 1 / 7.
expected['G'] = 1 / 7.
expected['T'] = 2 / 7.
self.assertEqual(self.b1.k_word_frequencies(1, overlapping=True),
expected)
expected = defaultdict(int)
expected['GAT'] = 1 / 5.
expected['ATT'] = 1 / 5.
expected['TTA'] = 1 / 5.
expected['TAC'] = 1 / 5.
expected['ACA'] = 1 / 5.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=True),
expected)
# overlapping = False
expected = defaultdict(int)
expected['GAT'] = 1 / 2.
expected['TAC'] = 1 / 2.
self.assertEqual(self.b1.k_word_frequencies(3, overlapping=False),
expected)
expected = defaultdict(int)
expected['GATTACA'] = 1.0
self.assertEqual(self.b1.k_word_frequencies(7, overlapping=False),
expected)
expected = defaultdict(int)
empty = BiologicalSequence('')
self.assertEqual(empty.k_word_frequencies(1, overlapping=False),
expected)
def test_len(self):
self.assertEqual(len(self.b1), 7)
self.assertEqual(len(self.b2), 9)
self.assertEqual(len(self.b3), 4)
def test_repr(self):
self.assertEqual(repr(self.b1),
"<BiologicalSequence: GATTACA (length: 7)>")
self.assertEqual(repr(self.b6),
"<BiologicalSequence: ACGTACGTAC... (length: 12)>")
def test_reversed(self):
b1_reversed = reversed(self.b1)
for actual, expected in zip(b1_reversed, "ACATTAG"):
self.assertEqual(actual, expected)
self.assertRaises(StopIteration, lambda: next(b1_reversed))
def test_str(self):
self.assertEqual(str(self.b1), "GATTACA")
self.assertEqual(str(self.b2), "ACCGGTACC")
self.assertEqual(str(self.b3), "GREG")
def test_alphabet(self):
self.assertEqual(self.b1.alphabet(), set())
def test_description(self):
self.assertEqual(self.b1.description, "")
self.assertEqual(self.b2.description, "A test sequence")
self.assertEqual(self.b3.description, "A protein sequence")
def test_gap_alphabet(self):
self.assertEqual(self.b1.gap_alphabet(), set('-.'))
def test_id(self):
self.assertEqual(self.b1.id, "")
self.assertEqual(self.b2.id, "test-seq-2")
self.assertEqual(self.b3.id, "test-seq-3")
def test_count(self):
self.assertEqual(self.b1.count('A'), 3)
self.assertEqual(self.b1.count('T'), 2)
self.assertEqual(self.b1.count('TT'), 1)
def test_degap(self):
self.assertEqual(self.b1.degap(), self.b1)
self.assertEqual(self.b7.degap(), BiologicalSequence(''))
self.assertEqual(self.b8.degap(), BiologicalSequence('HELLO'))
def test_distance(self):
# note that test_hamming_distance covers default behavior more
# extensively
self.assertEqual(self.b1.distance(self.b1), 0.0)
self.assertEqual(self.b1.distance(BiologicalSequence('GATTACC')),
1. / 7)
def dumb_distance(x, y):
return 42
self.assertEqual(self.b1.distance(self.b1, distance_fn=dumb_distance),
42)
def test_fraction_diff(self):
self.assertEqual(self.b1.fraction_diff(self.b1), 0., 5)
self.assertEqual(self.b1.fraction_diff(BiologicalSequence('GATTACC')),
1. / 7., 5)
def test_fraction_same(self):
self.assertAlmostEqual(self.b1.fraction_same(self.b1), 1., 5)
self.assertAlmostEqual(
self.b1.fraction_same(BiologicalSequence('GATTACC')), 6. / 7., 5)
def test_gap_maps(self):
# in sequence with no gaps, the gap_maps are identical
self.assertEqual(self.b1.gap_maps(),
([0, 1, 2, 3, 4, 5, 6], [0, 1, 2, 3, 4, 5, 6]))
# in sequence with all gaps, the map of degapped to gapped is the empty
# list (bc its length is 0), and the map of gapped to degapped is all
# None
self.assertEqual(self.b7.gap_maps(),
([], [None, None, None, None, None, None]))
self.assertEqual(self.b8.gap_maps(),
([0, 1, 8, 9, 10],
[0, 1, None, None, None, None, None, None, 2, 3, 4]))
# example from the gap_maps doc string
self.assertEqual(
BiologicalSequence('-ACCGA-TA-').gap_maps(),
([1, 2, 3, 4, 5, 7, 8], [None, 0, 1, 2, 3, 4, None, 5, 6, None]))
def test_gap_vector(self):
self.assertEqual(self.b1.gap_vector(), [False] * len(self.b1))
self.assertEqual(self.b7.gap_vector(), [True] * len(self.b7))
self.assertEqual(self.b8.gap_vector(), [
False, False, True, True, True, True, True, True, False, False,
False
])
def test_unsupported_characters(self):
self.assertEqual(self.b1.unsupported_characters(), set('GATC'))
self.assertEqual(self.b7.unsupported_characters(), set())
def test_has_unsupported_characters(self):
self.assertTrue(self.b1.has_unsupported_characters())
self.assertFalse(self.b7.has_unsupported_characters())
def test_index(self):
""" index functions as expected """
self.assertEqual(self.b1.index('G'), 0)
self.assertEqual(self.b1.index('A'), 1)
self.assertEqual(self.b1.index('AC'), 4)
self.assertRaises(ValueError, self.b1.index, 'x')
def test_is_gap(self):
self.assertTrue(self.b1.is_gap('.'))
self.assertTrue(self.b1.is_gap('-'))
self.assertFalse(self.b1.is_gap('A'))
self.assertFalse(self.b1.is_gap('x'))
self.assertFalse(self.b1.is_gap(' '))
self.assertFalse(self.b1.is_gap(''))
def test_is_gapped(self):
self.assertFalse(self.b1.is_gapped())
self.assertFalse(self.b2.is_gapped())
self.assertTrue(self.b7.is_gapped())
self.assertTrue(self.b8.is_gapped())
def test_is_valid(self):
self.assertFalse(self.b1.is_valid())
self.assertTrue(self.b7.is_valid())
def test_to_fasta(self):
self.assertEqual(self.b1.to_fasta(), ">\nGATTACA\n")
self.assertEqual(self.b1.to_fasta(terminal_character=""), ">\nGATTACA")
self.assertEqual(self.b2.to_fasta(),
">test-seq-2 A test sequence\nACCGGTACC\n")
self.assertEqual(self.b3.to_fasta(),
">test-seq-3 A protein sequence\nGREG\n")
self.assertEqual(self.b4.to_fasta(), ">test-seq-4\nPRTEIN\n")
self.assertEqual(self.b5.to_fasta(), "> some description\nLLPRTEIN\n")
# alt parameters
self.assertEqual(self.b2.to_fasta(field_delimiter=":"),
">test-seq-2:A test sequence\nACCGGTACC\n")
self.assertEqual(self.b2.to_fasta(terminal_character="!"),
">test-seq-2 A test sequence\nACCGGTACC!")
self.assertEqual(
self.b2.to_fasta(field_delimiter=":", terminal_character="!"),
">test-seq-2:A test sequence\nACCGGTACC!")
def test_upper(self):
b = NucleotideSequence('GAt.ACa-', id='x', description='42')
expected = NucleotideSequence('GAT.ACA-', id='x', description='42')
self.assertEqual(b.upper(), expected)
def test_lower(self):
b = NucleotideSequence('GAt.ACa-', id='x', description='42')
expected = NucleotideSequence('gat.aca-', id='x', description='42')
self.assertEqual(b.lower(), expected)
def setUp(self):
self.b1 = BiologicalSequence('GATTACA')
self.b2 = BiologicalSequence('ACCGGTACC',
id="test-seq-2",
description="A test sequence")
self.b3 = BiologicalSequence('GREG',
id="test-seq-3",
description="A protein sequence")
self.b4 = BiologicalSequence('PRTEIN', id="test-seq-4")
self.b5 = BiologicalSequence('LLPRTEIN',
description="some description")
self.b6 = BiologicalSequence('ACGTACGTACGT')
self.b7 = BiologicalSequence('..--..')
self.b8 = BiologicalSequence('HE..--..LLO')
def test_degap(self):
self.assertEqual(self.b1.degap(), self.b1)
self.assertEqual(self.b7.degap(), BiologicalSequence(''))
self.assertEqual(self.b8.degap(), BiologicalSequence('HELLO'))
def fasta_from_alignment(aln, make_seqlabel=None, line_wrap=None, sort=True):
"""Returns a FASTA string given an alignment object
Parameters
----------
aln : Alignment, dict
alignment or dictionary where the keys are the sequence ids and
the values are the sequences themselves.
make_seqlabel : function, optional
callback function that takes the seq object and returns a label
``str``. If ``None`` is passed, the following attributes will try to be
retrieved in this order and the first to exist will be used:
``id``, ``Label`` or ``Name``. In any other case an integer
with the position of the sequence object will be used.
line_wrap : int, optional
line_wrap: a integer for maximum line width, if ``None`` is passed the
full sequence will be used.
sort : bool, optional
Whether or not the sequences should be sorted by their sequence
id, default value is ``True``.
Returns
-------
str
FASTA formatted string composed of the objects passed in via `seqs`.
See Also
--------
skbio.parse.sequences.parse_fasta
skbio.core.alignment.Alignment
Examples
--------
Formatting a sequence alignment object into a FASTA file.
>>> from skbio.core.alignment import Alignment
>>> from skbio.core.sequence import DNA
>>> from skbio.format.sequences import fasta_from_alignment
>>> seqs = [DNA("ACC--G-GGTA..", id="seq1"),
... DNA("TCC--G-GGCA..", id="seqs2")]
>>> a1 = Alignment(seqs)
>>> print fasta_from_alignment(a1)
>seq1
ACC--G-GGTA..
>seqs2
TCC--G-GGCA..
"""
# check if it's an Alignment object or a dictionary
if isinstance(aln, Alignment):
order = aln.ids()
else:
order = aln.keys()
if sort:
order = sorted(order)
ordered_seqs = []
for label in order:
seq = aln[label]
if isinstance(seq, str):
seq = BiologicalSequence(seq, label)
ordered_seqs.append(seq)
return fasta_from_sequences(ordered_seqs,
make_seqlabel=make_seqlabel,
line_wrap=line_wrap)
def test_fraction_same(self):
self.assertAlmostEqual(self.b1.fraction_same(self.b1), 1., 5)
self.assertAlmostEqual(
self.b1.fraction_same(BiologicalSequence('GATTACC')), 6. / 7., 5)
def test_fraction_diff(self):
self.assertEqual(self.b1.fraction_diff(self.b1), 0., 5)
self.assertEqual(self.b1.fraction_diff(BiologicalSequence('GATTACC')),
1. / 7., 5)
