def test_seq_equality(self):
model1 = ArraySequence("UCG", name="rna1", alphabet=RNA.alphabets.degen_gapped)
model2 = ArraySequence("UCG", name="rna1", alphabet=RNA.alphabets.degen_gapped)
# Shouldn't the above two sequences be equal?
self.assertEqual(model1, model2)
# string comparison is True
self.assertEqual(str(model1), str(model2))
def test_ArrayAlignment_without_moltype(self):
"""Expect MolType to be picked up from the sequences."""
m1 = ArraySequence("UCAG", alphabet=RNA.alphabets.degen_gapped, name="rna1")
m2 = ArraySequence("CCCR", alphabet=RNA.alphabets.degen_gapped, name="rna2")
da = ArrayAlignment([m1, m2])
exp_lines = [">rna1", "UCAG", ">rna2", "CCCR"]
self.assertEqual(str(da), "\n".join(exp_lines) + "\n")
def test_subset_positions_ArrayAlignment(self):
# because dict order volatile, need to grab the
# the index for ambig characters from the object
# The full data comes from these seqs
# 'UCAGGG'
# 'YCU-RG'
# 'CAA-NR'
get_index = RNA.alphabets.degen_gapped.index
G = get_index("-")
N = get_index("N")
R = get_index("R")
Y = get_index("Y")
full_data = array([[0, 1, 2, 3, 3, 3], [Y, 1, 0, G, R, 3],
[1, 2, 2, G, N, R]])
model1 = ArraySequence("UCG",
name="rna1",
alphabet=RNA.alphabets.degen_gapped)
model2 = ArraySequence("YCG",
name="rna2",
alphabet=RNA.alphabets.degen_gapped)
model3 = ArraySequence("CAR",
name="rna3",
alphabet=RNA.alphabets.degen_gapped)
sub_da = ArrayAlignment([model1, model2, model3],
moltype=RNA,
alphabet=RNA.alphabets.degen_gapped)
sub_data = array([[0, 1, 3], [Y, 1, 3], [1, 2, R]])
# First check some data
self.assertEqual(self.da.array_seqs, full_data)
self.assertEqual(self.da.array_positions, transpose(full_data))
self.assertEqual(sub_da.array_seqs, sub_data)
self.assertEqual(sub_da.array_positions, transpose(sub_data))
obs_sub_da_TP = self.da.take_positions([0, 1, 5])
obs_sub_da_SA = self.da.get_sub_alignment(pos=[0, 1, 5])
# When using the get_sub_alignment method the data is right
self.assertEqual(obs_sub_da_SA, sub_da)
self.assertNotEqual(obs_sub_da_SA, self.da)
self.assertEqual(obs_sub_da_SA.array_seqs, sub_data)
self.assertEqual(obs_sub_da_SA.array_positions, transpose(sub_data))
# For the take_positions method: Why does this work
self.assertEqual(obs_sub_da_TP, sub_da)
self.assertNotEqual(obs_sub_da_TP, self.da)
# If the data doesn't match?
self.assertEqual(obs_sub_da_TP.array_seqs, sub_data)
self.assertEqual(obs_sub_da_TP.array_positions, transpose(sub_data))
def test_seq_ungapping(self):
rna1 = RnaSequence("U-C-A-G-", name="rna1")
model1 = ArraySequence("U-C-A-G-",
name="rna1",
alphabet=RNA.alphabets.degen_gapped)
self.assertEqual(rna1, "U-C-A-G-")
self.assertEqual(rna1.degap(), "UCAG")
# check is produces the right string from the beginning
self.assertEqual(str(model1), "U-C-A-G-")
self.assertEqual(model1._data, [0, 4, 1, 4, 2, 4, 3, 4])
# ArraySequence should maybe have the same degap method as normal seq
self.assertEqual(str(model1.degap()), "UCAG")
def test_subset_seqs_ArrayAlignment(self):
model1 = ArraySequence("UCG", name="rna1", alphabet=RNA.alphabets.degen_gapped)
model2 = ArraySequence("YCG", name="rna2", alphabet=RNA.alphabets.degen_gapped)
model3 = ArraySequence("CAR", name="rna3", alphabet=RNA.alphabets.degen_gapped)
sub_da = ArrayAlignment(
[model1, model2, model3], moltype=RNA, alphabet=RNA.alphabets.degen_gapped
)
# take_seqs by name should have the same effect as
# get_sub_alignment by seq idx?
obs_sub_da_TS = self.da.take_seqs(["rna1"])
obs_sub_da_SA = self.da.get_sub_alignment(seqs=[0])
# These two are now the same. Fixed mapping of key to char array.
self.assertEqual(obs_sub_da_TS, obs_sub_da_SA)
self.assertEqual(str(obs_sub_da_TS), str(obs_sub_da_SA))
def setUp(self):
"""setUp method for all tests"""
# named sequences
self.rna1 = RnaSequence("UCAGGG", name="rna1")
self.rna2 = RnaSequence("YCU-RG", name="rna2")
self.rna3 = RnaSequence("CAA-NR", name="rna3")
self.model1 = ArraySequence("UCAGGG",
name="rna1",
alphabet=RNA.alphabets.degen_gapped)
self.model2 = ArraySequence("YCU-RG",
name="rna2",
alphabet=RNA.alphabets.degen_gapped)
self.model3 = ArraySequence("CAA-NR",
name="rna3",
alphabet=RNA.alphabets.degen_gapped)
self.aln = Alignment([self.rna1, self.rna2, self.rna3], moltype=RNA)
self.da = ArrayAlignment(
[self.model1, self.model2, self.model3],
moltype=RNA,
alphabet=RNA.alphabets.degen_gapped,
)
# seqs no name
self.nn_rna1 = RnaSequence("UCAGGG")
self.nn_rna2 = RnaSequence("YCU-RG")
self.nn_rna3 = RnaSequence("CAA-NR")
self.nn_model1 = ArraySequence("UCAGGG",
alphabet=RNA.alphabets.degen_gapped)
self.nn_model2 = ArraySequence("YCU-RG",
alphabet=RNA.alphabets.degen_gapped)
self.nn_model3 = ArraySequence("CAA-NR",
alphabet=RNA.alphabets.degen_gapped)
self.nn_aln = Alignment([self.nn_rna1, self.nn_rna2, self.nn_rna3],
moltype=RNA)
self.nn_da = ArrayAlignment(
[self.nn_model1, self.nn_model2, self.nn_model3],
moltype=RNA,
alphabet=RNA.alphabets.degen_gapped,
)
