def test_subset_positions_DenseAlignment(self):
model1 = ModelSequence('UCG', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
model2 = ModelSequence('YCG', Name='rna2',\
Alphabet=RNA.Alphabets.DegenGapped)
model3 = ModelSequence('CAR', Name='rna3',\
Alphabet=RNA.Alphabets.DegenGapped)
sub_da = DenseAlignment([model1, model2, model3],\
MolType=RNA, Alphabet=RNA.Alphabets.DegenGapped)
full_data = array([[0, 1, 2, 3, 3, 3], [15, 1, 0, 4, 12, 3],
[1, 2, 2, 4, 10, 12]])
sub_data = array([[0, 1, 3], [15, 1, 3], [1, 2, 12]])
# First check some data
self.assertEqual(self.da.ArraySeqs, full_data)
self.assertEqual(self.da.ArrayPositions, transpose(full_data))
self.assertEqual(sub_da.ArraySeqs, sub_data)
self.assertEqual(sub_da.ArrayPositions, transpose(sub_data))
obs_sub_da_TP = self.da.takePositions([0, 1, 5])
obs_sub_da_SA = self.da.getSubAlignment(pos=[0, 1, 5])
# When using the getSubAlignment 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.ArraySeqs, sub_data)
self.assertEqual(obs_sub_da_SA.ArrayPositions, transpose(sub_data))
# For the takePositions 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.ArraySeqs, sub_data)
self.assertEqual(obs_sub_da_TP.ArrayPositions, transpose(sub_data))
def test_seq_equality(self):
model1 = ModelSequence('UCG', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
model2 = ModelSequence('UCG', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
# Shouldn't the above two sequences be equal?
self.assertEqual(model1, model2)
# string comparison is True
self.assertEqual(str(model1), str(model2))
def test_DenseAlignment_without_moltype(self):
"""Expect MolType to be picked up from the sequences."""
m1 = ModelSequence('UCAG',Alphabet=RNA.Alphabets.DegenGapped,\
Name='rna1')
m2 = ModelSequence('CCCR',Alphabet=RNA.Alphabets.DegenGapped,\
Name='rna2')
da = DenseAlignment([m1, m2])
exp_lines = ['>rna1', 'UCAG', '>rna2', 'CCCR']
self.assertEqual(str(da), '\n'.join(exp_lines) + '\n')
def test_seq_ungapping(self):
rna1 = RnaSequence('U-C-A-G-', Name='rna1')
model1 = ModelSequence('U-C-A-G-', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
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])
# ModelSequence should maybe have the same degap method as normal Seq
self.assertEqual(str(model1.degap()), 'UCAG')
def test_seq_ungapping(self):
rna1 = RnaSequence('U-C-A-G-', Name='rna1')
model1 = ModelSequence('U-C-A-G-', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
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])
# ModelSequence should maybe have the same degap method as normal Seq
self.assertEqual(str(model1.degap()), 'UCAG')
def setUp(self):
"""setUp: set up method for all tests"""
self.rna1 = RnaSequence('UCAG-RYN-N', Name='rna1')
self.m1 = ModelSequence('UCAG-RYN-N', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
self.s1 = 'UCAG-RYN-N'
def test_extract_seqs(self):
"""extract_seqs: should handle different input formats"""
s1 = ">seq1\nACGUAGC\n>seq2\nGGUAGCG"
s2 = [">seq1", "ACGUAGC", ">seq2", "GGUAGCG"]
s3 = ['ACGUAGC', 'GGUAGCG']
s4 = [RnaSequence('ACGUAGC'), RnaSequence('GGUAGCG')]
m1 = ModelSequence('ACGUAGC', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
m2 = ModelSequence('GGUAGCG', Name='rna2',\
Alphabet=RNA.Alphabets.DegenGapped)
s5 = [m1, m2]
f = extract_seqs
self.assertEqual(f(s1), ['ACGUAGC', 'GGUAGCG'])
self.assertEqual(f(s2), ['ACGUAGC', 'GGUAGCG'])
self.assertEqual(f(s3), ['ACGUAGC', 'GGUAGCG'])
self.assertEqual(f(s4), ['ACGUAGC', 'GGUAGCG'])
self.assertEqual(f(s5), ['ACGUAGC', 'GGUAGCG'])
def test_subset_seqs_DenseAlignment(self):
model1 = ModelSequence('UCG', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
model2 = ModelSequence('YCG', Name='rna2',\
Alphabet=RNA.Alphabets.DegenGapped)
model3 = ModelSequence('CAR', Name='rna3',\
Alphabet=RNA.Alphabets.DegenGapped)
sub_da = DenseAlignment([model1, model2, model3],\
MolType=RNA, Alphabet=RNA.Alphabets.DegenGapped)
# takeSeqs by name should have the same effect as
# getSubAlignment by seq idx?
obs_sub_da_TS = self.da.takeSeqs(['rna1'])
obs_sub_da_SA = self.da.getSubAlignment(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 test_score_sequence_object(self):
"""score: should work correctly on Sequence object as input
"""
# DnaSequence object
ds = self.score1.score(DNA.Sequence("ATTCAC"),offset=0)
self.assertEqual(ds, [6,2,-3,0])
# ModelSequence object
ms = self.score1.score(ModelSequence("ATTCAC", Alphabet=DNA.Alphabet),\
offset=0)
self.assertEqual(ms, [6,2,-3,0])
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 = ModelSequence('UCAGGG', Name='rna1',\
Alphabet=RNA.Alphabets.DegenGapped)
self.model2 = ModelSequence('YCU-RG', Name='rna2',\
Alphabet=RNA.Alphabets.DegenGapped)
self.model3 = ModelSequence('CAA-NR', Name='rna3',\
Alphabet=RNA.Alphabets.DegenGapped)
self.aln = Alignment([self.rna1, self.rna2, self.rna3], MolType=RNA)
self.da = DenseAlignment([self.model1, self.model2, self.model3],\
MolType=RNA, Alphabet=RNA.Alphabets.DegenGapped)
# seqs no name
self.nn_rna1 = RnaSequence('UCAGGG')
self.nn_rna2 = RnaSequence('YCU-RG')
self.nn_rna3 = RnaSequence('CAA-NR')
self.nn_model1 = ModelSequence('UCAGGG',\
Alphabet=RNA.Alphabets.DegenGapped)
self.nn_model2 = ModelSequence('YCU-RG',\
Alphabet=RNA.Alphabets.DegenGapped)
self.nn_model3 = ModelSequence('CAA-NR',\
Alphabet=RNA.Alphabets.DegenGapped)
self.nn_aln = Alignment([self.nn_rna1, self.nn_rna2, self.nn_rna3],\
MolType=RNA)
self.nn_da = DenseAlignment([self.nn_model1, self.nn_model2,\
self.nn_model3], MolType=RNA, Alphabet=RNA.Alphabets.DegenGapped)