def test_phenotype_IO(self):
'''Test basic functionalities of phenotype IO methods'''
p1 = phenotype.read(SMALL_JSON_PLATE, 'pm-json')
p2 = next(phenotype.parse(SMALL_CSV_PLATES, 'pm-csv'))
handle = StringIO()
c = phenotype.write([p1, p2], handle, 'pm-json')
self.assertEqual(c, 2)
handle.flush()
handle.seek(0)
# Now ready to read back from the handle...
try:
records = list(phenotype.parse(handle, 'pm-json'))
except ValueError as e:
# This is BAD. We can't read our own output.
# I want to see the output when called from the test harness,
# run_tests.py (which can be funny about new lines on Windows)
handle.seek(0)
raise ValueError("%s\n\n%s\n\n%s"
% (str(e), repr(handle.read()), repr(records)))
self.assertEqual(p1, records[0])
handle.close()
handle = StringIO()
self.assertRaises(TypeError, phenotype.write, p1, handle, 1)
self.assertRaises(ValueError, phenotype.write, p1, handle, 'PM-JSON')
self.assertRaises(ValueError, phenotype.write, p1, handle, 'pm-csv')
handle.close()
def test_phenotype_IO(self):
"""Test basic functionalities of phenotype IO methods."""
p1 = phenotype.read(SMALL_JSON_PLATE, 'pm-json')
p2 = next(phenotype.parse(SMALL_CSV_PLATES, 'pm-csv'))
handle = StringIO()
c = phenotype.write([p1, p2], handle, 'pm-json')
self.assertEqual(c, 2)
handle.flush()
handle.seek(0)
# Now ready to read back from the handle...
try:
records = list(phenotype.parse(handle, 'pm-json'))
except ValueError as e:
# This is BAD. We can't read our own output.
# I want to see the output when called from the test harness,
# run_tests.py (which can be funny about new lines on Windows)
handle.seek(0)
raise ValueError("%s\n\n%s\n\n%s"
% (str(e), repr(handle.read()), repr(records)))
self.assertEqual(p1, records[0])
handle.close()
handle = StringIO()
self.assertRaises(TypeError, phenotype.write, p1, handle, 1)
self.assertRaises(ValueError, phenotype.write, p1, handle, 'PM-JSON')
self.assertRaises(ValueError, phenotype.write, p1, handle, 'pm-csv')
handle.close()
def __str__(self):
"""Create a string representation of the MarkovModel object."""
from Bio._py3k import StringIO
handle = StringIO()
save(self, handle)
handle.seek(0)
return handle.read()
def __str__(self):
"""Create a string representation of the MarkovModel object."""
from Bio._py3k import StringIO
handle = StringIO()
save(self, handle)
handle.seek(0)
return handle.read()
def test_write_alignment(self):
# Default causes no interleave (columns <= 1000)
records = [SeqRecord(Seq("ATGCTGCTGA" * 90, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a)
handle.seek(0)
data = handle.read()
self.assertIn("ATGCTGCTGA" * 90, data)
# Default causes interleave (columns > 1000)
records = [SeqRecord(Seq("ATGCTGCTGA" * 110, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a)
handle.seek(0)
data = handle.read()
self.assertNotIn("ATGCTGCTGA" * 90, data)
self.assertIn("ATGCTGCTGA" * 7, data)
# Override interleave: True
records = [SeqRecord(Seq("ATGCTGCTGA" * 9, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a, interleave=True)
handle.seek(0)
data = handle.read()
self.assertNotIn("ATGCTGCTGA" * 9, data)
self.assertIn("ATGCTGCTGA" * 7, data)
# Override interleave: False
records = [SeqRecord(Seq("ATGCTGCTGA" * 110, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a, interleave=False)
handle.seek(0)
data = handle.read()
self.assertIn("ATGCTGCTGA" * 110, data)
def test_write_alignment(self):
# Default causes no interleave (columns <= 1000)
records = [SeqRecord(Seq("ATGCTGCTGA" * 90, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a)
handle.seek(0)
data = handle.read()
self.assertIn("ATGCTGCTGA" * 90, data)
# Default causes interleave (columns > 1000)
records = [SeqRecord(Seq("ATGCTGCTGA" * 110, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a)
handle.seek(0)
data = handle.read()
self.assertNotIn("ATGCTGCTGA" * 90, data)
self.assertIn("ATGCTGCTGA" * 7, data)
# Override interleave: True
records = [SeqRecord(Seq("ATGCTGCTGA" * 9, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a, interleave=True)
handle.seek(0)
data = handle.read()
self.assertNotIn("ATGCTGCTGA" * 9, data)
self.assertIn("ATGCTGCTGA" * 7, data)
# Override interleave: False
records = [SeqRecord(Seq("ATGCTGCTGA" * 110, alphabet=ambiguous_dna), id=_id) for _id in ["foo", "bar", "baz"]]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_alignment(a, interleave=False)
handle.seek(0)
data = handle.read()
self.assertIn("ATGCTGCTGA" * 110, data)
def test_multiple_output(self):
records = [
SeqRecord(Seq("ATGCTGCTGAT", alphabet=ambiguous_dna), id="foo"),
SeqRecord(Seq("ATGCTGCAGAT", alphabet=ambiguous_dna), id="bar"),
SeqRecord(Seq("ATGCTGCGGAT", alphabet=ambiguous_dna), id="baz")
]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_file([a])
handle.seek(0)
data = handle.read()
self.assertTrue(data.startswith("#NEXUS\nbegin data;\n"), data)
self.assertTrue(data.endswith("end;\n"), data)
handle = StringIO()
with self.assertRaises(ValueError):
NexusWriter(handle).write_file([a, a])
def test_multiple_output(self):
records = [SeqRecord(Seq("ATGCTGCTGAT", alphabet=ambiguous_dna), id="foo"),
SeqRecord(Seq("ATGCTGCAGAT", alphabet=ambiguous_dna), id="bar"),
SeqRecord(Seq("ATGCTGCGGAT", alphabet=ambiguous_dna), id="baz")]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_file([a])
handle.seek(0)
data = handle.read()
self.assertTrue(data.startswith("#NEXUS\nbegin data;\n"), data)
self.assertTrue(data.endswith("end;\n"), data)
handle = StringIO()
try:
NexusWriter(handle).write_file([a, a])
assert False, "Should have rejected more than one alignment!"
except ValueError:
pass
def test_empty_file(self):
print("Reading an empty file")
assert 0 == len(list(NexusIterator(StringIO())))
print("Done")
print("")
print("Writing...")
handle = StringIO()
NexusWriter(handle).write_file([a])
handle.seek(0)
print(handle.read())
handle = StringIO()
try:
NexusWriter(handle).write_file([a, a])
assert False, "Should have rejected more than one alignment!"
except ValueError:
pass
def test_multiple_output(self):
records = [SeqRecord(Seq("ATGCTGCTGAT", alphabet=ambiguous_dna), id="foo"),
SeqRecord(Seq("ATGCTGCAGAT", alphabet=ambiguous_dna), id="bar"),
SeqRecord(Seq("ATGCTGCGGAT", alphabet=ambiguous_dna), id="baz")]
a = MultipleSeqAlignment(records, alphabet=ambiguous_dna)
handle = StringIO()
NexusWriter(handle).write_file([a])
handle.seek(0)
data = handle.read()
self.assertTrue(data.startswith("#NEXUS\nbegin data;\n"), data)
self.assertTrue(data.endswith("end;\n"), data)
handle = StringIO()
try:
NexusWriter(handle).write_file([a, a])
assert False, "Should have rejected more than one alignment!"
except ValueError:
pass
def test_empty_file(self):
print("Reading an empty file")
assert 0 == len(list(NexusIterator(StringIO())))
print("Done")
print("")
print("Writing...")
handle = StringIO()
NexusWriter(handle).write_file([a])
handle.seek(0)
print(handle.read())
handle = StringIO()
try:
NexusWriter(handle).write_file([a, a])
assert False, "Should have rejected more than one alignment!"
except ValueError:
pass
def check_simple_write_read(alignments, indent=" "):
# print(indent+"Checking we can write and then read back these alignments")
for format in test_write_read_align_with_seq_count:
records_per_alignment = len(alignments[0])
for a in alignments:
if records_per_alignment != len(a):
records_per_alignment = None
# Can we expect this format to work?
if not records_per_alignment \
and format not in test_write_read_alignment_formats:
continue
print(indent + "Checking can write/read as '%s' format" % format)
# Going to write to a handle...
handle = StringIO()
try:
c = AlignIO.write(alignments, handle=handle, format=format)
assert c == len(alignments)
except ValueError as e:
# This is often expected to happen, for example when we try and
# write sequences of different lengths to an alignment file.
print(indent + "Failed: %s" % str(e))
# Carry on to the next format:
continue
# First, try with the seq_count
if records_per_alignment:
handle.flush()
handle.seek(0)
try:
alignments2 = list(
AlignIO.parse(handle=handle,
format=format,
seq_count=records_per_alignment))
except ValueError as e:
# This is BAD. We can't read our own output.
# I want to see the output when called from the test harness,
# run_tests.py (which can be funny about new lines on Windows)
handle.seek(0)
raise ValueError(
"%s\n\n%s\n\n%s" %
(str(e), repr(handle.read()), repr(alignments2)))
simple_alignment_comparison(alignments, alignments2, format)
if format in test_write_read_alignment_formats:
# Don't need the seq_count
handle.flush()
handle.seek(0)
try:
alignments2 = list(AlignIO.parse(handle=handle, format=format))
except ValueError as e:
# This is BAD. We can't read our own output.
# I want to see the output when called from the test harness,
# run_tests.py (which can be funny about new lines on Windows)
handle.seek(0)
raise ValueError(
"%s\n\n%s\n\n%s" %
(str(e), repr(handle.read()), repr(alignments2)))
simple_alignment_comparison(alignments, alignments2, format)
if len(alignments) > 1:
# Try writing just one Alignment (not a list)
handle = StringIO()
AlignIO.write(alignments[0:1], handle, format)
assert handle.getvalue() == alignments[0].format(format)
if t_format != "nexus":
# Hack for bug 2535
pssm = summary.pos_specific_score_matrix()
rep_dict = summary.replacement_dictionary()
try:
info_content = summary.information_content()
except ValueError as err:
if str(
err
) != "Error in alphabet: not Nucleotide or Protein, supply expected frequencies":
raise err
if t_count == 1 and t_format not in ["nexus", "emboss", "fasta-m10"]:
# print(" Trying to read a triple concatenation of the input file")
with open(t_filename, "r") as handle:
data = handle.read()
handle = StringIO()
handle.write(data + "\n\n" + data + "\n\n" + data)
handle.seek(0)
assert len(
list(AlignIO.parse(handle=handle, format=t_format,
seq_count=t_per))) == 3
handle.close()
# Some alignment file formats have magic characters which mean
# use the letter in this position in the first sequence.
# They should all have been converted by the parser, but if
# not reversing the record order might expose an error. Maybe.
alignments.reverse()
check_simple_write_read(alignments)
def check_simple_write_read(alignments, indent=" "):
# print(indent+"Checking we can write and then read back these alignments")
for format in test_write_read_align_with_seq_count:
records_per_alignment = len(alignments[0])
for a in alignments:
if records_per_alignment != len(a):
records_per_alignment = None
# Can we expect this format to work?
if not records_per_alignment \
and format not in test_write_read_alignment_formats:
continue
print(indent + "Checking can write/read as '%s' format" % format)
# Going to write to a handle...
handle = StringIO()
try:
c = AlignIO.write(alignments, handle=handle, format=format)
assert c == len(alignments)
except ValueError as e:
# This is often expected to happen, for example when we try and
# write sequences of different lengths to an alignment file.
print(indent + "Failed: %s" % str(e))
# Carry on to the next format:
continue
# First, try with the seq_count
if records_per_alignment:
handle.flush()
handle.seek(0)
try:
alignments2 = list(AlignIO.parse(handle=handle, format=format,
seq_count=records_per_alignment))
except ValueError as e:
# This is BAD. We can't read our own output.
# I want to see the output when called from the test harness,
# run_tests.py (which can be funny about new lines on Windows)
handle.seek(0)
raise ValueError("%s\n\n%s\n\n%s"
% (str(e), repr(handle.read()), repr(alignments2)))
simple_alignment_comparison(alignments, alignments2, format)
if format in test_write_read_alignment_formats:
# Don't need the seq_count
handle.flush()
handle.seek(0)
try:
alignments2 = list(AlignIO.parse(handle=handle, format=format))
except ValueError as e:
# This is BAD. We can't read our own output.
# I want to see the output when called from the test harness,
# run_tests.py (which can be funny about new lines on Windows)
handle.seek(0)
raise ValueError("%s\n\n%s\n\n%s"
% (str(e), repr(handle.read()), repr(alignments2)))
simple_alignment_comparison(alignments, alignments2, format)
if len(alignments) > 1:
# Try writing just one Alignment (not a list)
handle = StringIO()
SeqIO.write(alignments[0], handle, format)
assert handle.getvalue() == alignments[0].format(format)
# gap_consensus = summary.gap_consensus()
if t_format != "nexus":
# Hack for bug 2535
pssm = summary.pos_specific_score_matrix()
rep_dict = summary.replacement_dictionary()
try:
info_content = summary.information_content()
except ValueError as e:
if str(e) != "Error in alphabet: not Nucleotide or Protein, supply expected frequencies":
raise e
pass
if t_count == 1 and t_format not in ["nexus", "emboss", "fasta-m10"]:
# print(" Trying to read a triple concatenation of the input file")
with open(t_filename, "r") as handle:
data = handle.read()
handle = StringIO()
handle.write(data + "\n\n" + data + "\n\n" + data)
handle.seek(0)
assert 3 == len(list(AlignIO.parse(handle=handle, format=t_format, seq_count=t_per)))
handle.close()
# Some alignment file formats have magic characters which mean
# use the letter in this position in the first sequence.
# They should all have been converted by the parser, but if
# not reversing the record order might expose an error. Maybe.
alignments.reverse()
check_simple_write_read(alignments)
print("Finished tested reading files")
CYS1_DICDI -----MKVIL LFVLAVFTVF VSS------- --------RG IPPEEQ----
ALEU_HORVU MAHARVLLLA LAVLATAAVA VASSSSFADS NPIRPVTDRA ASTLESAVLG
CATH_HUMAN ------MWAT LPLLCAGAWL LGV------- -PVCGAAELS VNSLEK----
CYS1_DICDI -----MKVIL LFVLAVFTVF VSS------- --------RG IPPEEQ---X
;
end;
""")
for a in NexusIterator(handle):
print(a)
for r in a:
print("%r %s %s" % (r.seq, r.name, r.id))
print("Done")
print("")
print("Reading an empty file")
assert 0 == len(list(NexusIterator(StringIO())))
print("Done")
print("")
print("Writing...")
handle = StringIO()
NexusWriter(handle).write_file([a])
handle.seek(0)
print(handle.read())
handle = StringIO()
try:
NexusWriter(handle).write_file([a, a])
assert False, "Should have rejected more than one alignment!"
except ValueError:
pass
def __str__(self):
from Bio._py3k import StringIO
handle = StringIO()
save(self, handle)
handle.seek(0)
return handle.read()
CYS1_DICDI -----MKVIL LFVLAVFTVF VSS------- --------RG IPPEEQ----
ALEU_HORVU MAHARVLLLA LAVLATAAVA VASSSSFADS NPIRPVTDRA ASTLESAVLG
CATH_HUMAN ------MWAT LPLLCAGAWL LGV------- -PVCGAAELS VNSLEK----
CYS1_DICDI -----MKVIL LFVLAVFTVF VSS------- --------RG IPPEEQ---X
;
end;
""")
for a in NexusIterator(handle):
print(a)
for r in a:
print("%r %s %s" % (r.seq, r.name, r.id))
print("Done")
print("")
print("Reading an empty file")
assert 0 == len(list(NexusIterator(StringIO())))
print("Done")
print("")
print("Writing...")
handle = StringIO()
NexusWriter(handle).write_file([a])
handle.seek(0)
print(handle.read())
handle = StringIO()
try:
NexusWriter(handle).write_file([a, a])
assert False, "Should have rejected more than one alignment!"
except ValueError:
pass
