def test_init(self):
"""JointEnumeration init should work as expected"""
# should work for alphabet object
a = JointEnumeration([DnaBases, RnaBases])
self.assertEqual(len(a), 16)
self.assertEqual(a.shape, (4, 4))
self.assertEqual(a[0], ("T", "U"))
self.assertEqual(a[-1], ("G", "G"))
assert_equal(a._sub_enum_factors, array([[4], [1]]))
# should work for arbitrary sequences
a = JointEnumeration(["TCAG", "UCAG"])
self.assertEqual(len(a), 16)
self.assertEqual(a[0], ("T", "U"))
self.assertEqual(a[-1], ("G", "G"))
assert_equal(a._sub_enum_factors, array([[4], [1]]))
# should work for different length sequences
a = JointEnumeration(["TCA", "UCAG"])
self.assertEqual(a.shape, (3, 4))
self.assertEqual(len(a), 12)
self.assertEqual(a[0], ("T", "U"))
self.assertEqual(a[-1], ("A", "G"))
assert_equal(a._sub_enum_factors, array([[4],
[1]])) # note: _not_ [3,1]
def test_to_string(self):
"""CharAlphabet to_string should convert an input array to string"""
r = CharAlphabet("UCAG")
self.assertEqual(r.to_string(array([[0, 0, 1], [0, 3, 2]], "B")), "UUC\nUGA")
# should work with single seq
self.assertEqual(r.to_string(array([[0, 0, 1, 0, 3, 2]], "B")), "UUCUGA")
# should work with single seq
self.assertEqual(r.to_string(array([0, 0, 1, 0, 3, 2], "B")), "UUCUGA")
# should work with empty seq
self.assertEqual(r.to_string(array([], "B")), "")
def test_init(self):
"""CharAlphabet init should make correct translation tables"""
r = CharAlphabet("UCAG")
i2c, c2i = r._indices_nums_to_chars, r._chars_to_indices
s = array([0, 0, 1, 0, 3, 2], "b").tostring()
self.assertEqual(s.translate(i2c), b"UUCUGA")
self.assertEqual("UUCUGA".translate(c2i), "\000\000\001\000\003\002")
def test_unpack_arrays(self):
"""JointEnumeration unpack_arrays should return correct arrays."""
a = JointEnumeration(["xyz", "abcd", "ef"])
v = [7, 15, 18, 0]
result = a.unpack_arrays(v)
self.assertEqual(result,
array([[0, 1, 2, 0], [3, 3, 1, 0], [1, 1, 0, 0]]))
def test_pack_arrays(self):
"""JointEnumeration pack_arrays should return correct array."""
a = JointEnumeration(["xyz", "abcd", "ef"])
v = [[0, 1, 2, 0], [3, 3, 1, 0], [1, 1, 0, 0]]
result = a.pack_arrays(v)
assert_equal(result, array([7, 15, 18, 0]))
def test_from_seq_to_array(self):
"""convert a sequence into indices"""
dna = get_moltype("dna")
seq = dna.make_seq("ACGG")
got = dna.alphabet.from_seq_to_array(seq)
assert_equal(got, array([dna.alphabet.index(b) for b in seq]))
def test_to_chars(self):
"""CharAlphabet to_chars should convert an input array to chars"""
r = CharAlphabet("UCAG")
c = r.to_chars(array([[0, 0, 1], [0, 3, 2]], "B"))
assert_equal(c, array(["UUC", "UGA"], "c"))
def test_from_array(self):
"""CharAlphabet from_array should return correct array"""
r = CharAlphabet("UCAG")
got = r.from_array(array(["UUC", "UGA"], "c"))
assert_equal(got, array([[0, 0, 1], [0, 3, 2]], "B"))
def test_from_string(self):
"""CharAlphabet from_string should return correct array"""
r = CharAlphabet("UCAG")
assert_equal(r.from_string("UUCUGA"), array([0, 0, 1, 0, 3, 2], "B"))
def test_counts(self):
"""Enumeration counts should count freqs in array"""
a = DnaBases
f = array([[0, 0, 1, 0, 0, 3]])
assert_equal(a.counts(f), array([4, 1, 0, 1]))
# check that it works with byte array
f = array([[0, 0, 1, 0, 0, 3]], "B")
assert_equal(a.counts(f), array([4, 1, 0, 1]))
# should ignore out-of-bounds items
g = [0, 4]
assert_equal(a.counts(g), array([1, 0, 0, 0]))
# make sure it works for long sequences, i.e. no wraparound at 255
h = [0, 3] * 70000
assert_equal(a.counts(h), array([70000, 0, 0, 70000]))
h2 = array(h).astype("B")
assert_equal(a.counts(h2), array([70000, 0, 0, 70000]))
i = array([0, 3] * 75000)
assert_equal(a.counts(i), array([75000, 0, 0, 75000]))
# make sure it works for long _binary_ sequences, e.g. the results
# of array comparisons.
a = array([0, 1, 2, 3] * 10000)
b = array([0, 0, 0, 0] * 10000)
same = a == b
