def do_analysis_helper(labels, element_lists, w):
"""
Chop up the rows of data.
Yield lines of text to be displayed in an html pre tag.
@param labels: row labels to be left justified
@param element_lists: data rows where each element is a letter or a span
@param w: the width; the number of elements allowed per page row
"""
if len(set(len(element_list) for element_list in element_lists)) != 1:
msg = 'each element list should have the same nonzero length'
raise ValueError(msg)
label_width = max(len(label) for label in labels) + 1
chopped_element_lists = [
list(iterutils.chopped(element_list, w))
for element_list in element_lists
]
page_rows = zip(*chopped_element_lists)
for i, page_row in enumerate(page_rows):
header = ''
header += ' ' * label_width
header += Monospace.get_ruler_line(i * w + 1, i * w + len(page_row[0]))
yield header
for label, element_list in zip(labels, page_row):
justified_label = label.ljust(label_width)
yield ''.join([justified_label] + list(element_list))
if i < len(page_rows) - 1:
yield ''
def do_analysis_helper(labels, element_lists, w):
"""
Chop up the rows of data.
Yield lines of text to be displayed in an html pre tag.
@param labels: row labels to be left justified
@param element_lists: data rows where each element is a letter or a span
@param w: the width; the number of elements allowed per page row
"""
if len(set(len(element_list) for element_list in element_lists)) != 1:
msg = 'each element list should have the same nonzero length'
raise ValueError(msg)
label_width = max(len(label) for label in labels) + 1
chopped_element_lists = [list(iterutils.chopped(element_list, w))
for element_list in element_lists]
page_rows = zip(*chopped_element_lists)
for i, page_row in enumerate(page_rows):
header = ''
header += ' ' * label_width
header += Monospace.get_ruler_line(i*w + 1, i*w + len(page_row[0]))
yield header
for label, element_list in zip(labels, page_row):
justified_label = label.ljust(label_width)
yield ''.join([justified_label] + list(element_list))
if i < len(page_rows) - 1:
yield ''
def _init_codon_to_count(self):
# parse the raw string
codon_pattern = r'([ACGU][ACGU][ACGU])(.*)\((.*)\)'
line_pattern = '.*'.join([codon_pattern] * 4)
token_lists = []
for line in StringIO(_raw_human_codon_count_string):
line = line.strip()
if line:
m = re.search(line_pattern, line)
token_list = [x.strip() for x in m.groups()]
assert len(token_list) == 12
token_lists.append(token_list)
assert len(token_lists) == 16
# write the dictionary using the tokens
for token_list in token_lists:
for token in token_list:
assert type(token) == str
for codon, per_thousand, count in iterutils.chopped(token_list, 3):
# validate the codon
assert len(codon) == 3
assert set(codon) <= set('ACGU')
# use the dna sense codon instead of the rna codon
codon = codon.replace('U', 'T')
# validate the count
assert set(count) <= set('0123456789')
count = int(count)
assert count > 0
self._codon_to_count[codon] = count
assert len(self._codon_to_count) == 64
def get_fasta_sequence(self, header, ncols=60):
"""
@param ncols: the maximum number of residues per line
@return: a string representing the header and sequence in fasta format
"""
header_to_sequence = dict(zip(self.headers, self.sequences))
sequence = header_to_sequence[header]
arr = []
arr.append('>' + header)
print sequence
arr.append('\n'.join(iterutils.chopped(sequence, ncols)))
return '\n'.join(arr)
def get_alignment_string_non_interleaved(alignment):
"""
@param alignment: a fasta alignment object
@return: a non interleaved phylip alignment string
"""
out = StringIO()
# print the number of sequences and the length of each sequence
print >> out, len(alignment.headers), len(alignment.sequences[0])
# print each sequence
for header, sequence in zip(alignment.headers, alignment.sequences):
print >> out, header
for segment in iterutils.chopped(sequence, 60):
print >> out, segment
return out.getvalue().strip()
def get_alignment_string_non_interleaved(alignment):
"""
@param alignment: a fasta alignment object
@return: a non interleaved phylip alignment string
"""
out = StringIO()
# print the number of sequences and the length of each sequence
print >> out, len(alignment.headers), len(alignment.sequences[0])
# print each sequence
for header, sequence in zip(alignment.headers, alignment.sequences):
print >> out, header
for segment in iterutils.chopped(sequence, 60):
print >> out, segment
return out.getvalue().strip()
def __init__(self, lines):
"""
@param lines: lines of a fasta file
"""
if not lines:
raise AlignmentError('no fasta lines were provided')
header_sequence_pairs = list(gen_header_sequence_pairs(lines))
self.headers = [header for header, sequence in header_sequence_pairs]
nucleotide_sequences = [
seq.upper() for header, seq in header_sequence_pairs
]
if not nucleotide_sequences:
raise AlignmentError('no nucleotide sequences were found')
for header in self.headers:
if not header:
raise AlignmentError('each sequence should have a header')
if len(set(len(sequence) for sequence in nucleotide_sequences)) != 1:
raise AlignmentError('not all sequences are the same length')
for seq in nucleotide_sequences:
invalid_states = set(seq) - set('ACGT-')
if invalid_states:
example_invalid_state = invalid_states.pop()
raise CodonAlignmentError('invalid nucleotide: %s' %
example_invalid_state)
nucleotide_columns = zip(*nucleotide_sequences)
if len(nucleotide_columns) % 3 != 0:
raise CodonAlignmentError(
'the number of aligned nucleotide columns should be a multiple of three'
)
gappy_codon_sequences = [
list(iterutils.chopped(seq, 3)) for seq in nucleotide_sequences
]
if not gappy_codon_sequences:
raise CodonAlignmentError('no codon sequences were found')
observed_gappy_codons = set(
itertools.chain.from_iterable(gappy_codon_sequences))
valid_gappy_codons = set(list(Codon.g_non_stop_codons) + ['---'])
invalid_gappy_codons = observed_gappy_codons - valid_gappy_codons
if invalid_gappy_codons:
example_invalid_codon = invalid_gappy_codons.pop()
raise CodonAlignmentError('invalid codon: %s' %
example_invalid_codon)
self.columns = [
col for col in zip(*gappy_codon_sequences) if '---' not in col
]
if not self.columns:
raise CodonAlignmentError('no ungapped codon columns were found')
self.sequences = zip(*self.columns)
def get_alignment_string_interleaved(alignment):
"""
@param alignment: a fasta alignment object
@return: an interleaved phylip alignment string
"""
chopped_sequences = [
iterutils.chopped(seq, 60) for seq in alignment.sequences]
bands = zip(*chopped_sequences)
out = StringIO()
print >> out, len(alignment.headers), len(alignment.sequences[0])
lengths = [9] + [len(header) for header in alignment.headers]
n = max(lengths)
for header, segment in zip(alignment.headers, bands[0]):
aligned_header = header.ljust(n)
print >> out, '%s %s' % (aligned_header, segment)
print >> out, ''
for band in bands[1:]:
for segment in band:
print >> out, segment
print >> out, ''
return out.getvalue().strip()
def get_alignment_string_interleaved(alignment):
"""
@param alignment: a fasta alignment object
@return: an interleaved phylip alignment string
"""
chopped_sequences = [
iterutils.chopped(seq, 60) for seq in alignment.sequences
]
bands = zip(*chopped_sequences)
out = StringIO()
print >> out, len(alignment.headers), len(alignment.sequences[0])
lengths = [9] + [len(header) for header in alignment.headers]
n = max(lengths)
for header, segment in zip(alignment.headers, bands[0]):
aligned_header = header.ljust(n)
print >> out, '%s %s' % (aligned_header, segment)
print >> out, ''
for band in bands[1:]:
for segment in band:
print >> out, segment
print >> out, ''
return out.getvalue().strip()
def load(self, lines):
"""
@param lines: lines of nexus data
"""
# get the taxa, tree, and character lines
taxa_lines = []
tree_lines = []
character_lines = []
current_array = None
for line in iterutils.stripped_lines(lines):
# Ignore an entire line that is a comment.
# Nested comments and multi-line comments
# are not correctly processed here.
if line.startswith('[') and line.endswith(']'):
self.add_comment(line[1:-1])
continue
tokens = line.upper().split()
if tokens == ['BEGIN', 'TAXA;']:
current_array = taxa_lines
elif tokens == ['BEGIN', 'TREES;']:
current_array = tree_lines
elif tokens == ['BEGIN', 'CHARACTERS;']:
current_array = character_lines
elif tokens == ['END;']:
current_array = None
elif current_array is not None:
current_array.append(line)
# assert that tree lines and character lines are present
if not tree_lines:
raise NexusError('TREES was not found')
if not character_lines:
raise NexusError('CHARACTERS was not found')
# read the newick tree string
nexus_tree_string = ''.join(tree_lines)
if nexus_tree_string.count(';') != 1:
raise NexusError(
'expected exactly one semicolon in the nexus TREES block')
if nexus_tree_string.count('=') != 1:
raise NexusError(
'expected exactly one equals sign in the nexus TREES block')
offset = nexus_tree_string.find('=')
newick_string = nexus_tree_string[offset + 1:]
self.tree = Newick.parse(newick_string, Newick.NewickTree)
# read the alignment matrix
arr = []
found_matrix = False
for line in character_lines:
if line.upper().startswith('DIMENSIONS'):
continue
if line.upper().startswith('FORMAT'):
continue
if line.upper().startswith('MATRIX'):
found_matrix = True
continue
if found_matrix:
arr.append(line.replace(';', ' '))
if not arr:
raise NexusError('no alignment was found')
tokens = ' '.join(arr).split()
if len(tokens) % 2 != 0:
raise NexusError(
'expected the alignment to be a list of (taxon, sequence) pairs'
)
alignment_out = StringIO()
for header, sequence in iterutils.chopped(tokens, 2):
sequence = sequence.upper()
unexpected_letters = set(sequence) - set('ACGT')
if unexpected_letters:
raise NexusError('unexpected sequence character(s): %s' %
list(unexpected_letters))
print >> alignment_out, '>%s' % header
print >> alignment_out, sequence
alignment_string = alignment_out.getvalue()
self.alignment = Fasta.Alignment(StringIO(alignment_string))
def load(self, lines):
"""
@param lines: lines of nexus data
"""
# get the taxa, tree, and character lines
taxa_lines = []
tree_lines = []
character_lines = []
current_array = None
for line in iterutils.stripped_lines(lines):
# Ignore an entire line that is a comment.
# Nested comments and multi-line comments
# are not correctly processed here.
if line.startswith('[') and line.endswith(']'):
self.add_comment(line[1:-1])
continue
tokens = line.upper().split()
if tokens == ['BEGIN', 'TAXA;']:
current_array = taxa_lines
elif tokens == ['BEGIN', 'TREES;']:
current_array = tree_lines
elif tokens == ['BEGIN', 'CHARACTERS;']:
current_array = character_lines
elif tokens == ['END;']:
current_array = None
elif current_array is not None:
current_array.append(line)
# assert that tree lines and character lines are present
if not tree_lines:
raise NexusError('TREES was not found')
if not character_lines:
raise NexusError('CHARACTERS was not found')
# read the newick tree string
nexus_tree_string = ''.join(tree_lines)
if nexus_tree_string.count(';') != 1:
raise NexusError('expected exactly one semicolon in the nexus TREES block')
if nexus_tree_string.count('=') != 1:
raise NexusError('expected exactly one equals sign in the nexus TREES block')
offset = nexus_tree_string.find('=')
newick_string = nexus_tree_string[offset+1:]
self.tree = Newick.parse(newick_string, Newick.NewickTree)
# read the alignment matrix
arr = []
found_matrix = False
for line in character_lines:
if line.upper().startswith('DIMENSIONS'):
continue
if line.upper().startswith('FORMAT'):
continue
if line.upper().startswith('MATRIX'):
found_matrix = True
continue
if found_matrix:
arr.append(line.replace(';', ' '))
if not arr:
raise NexusError('no alignment was found')
tokens = ' '.join(arr).split()
if len(tokens) % 2 != 0:
raise NexusError('expected the alignment to be a list of (taxon, sequence) pairs')
alignment_out = StringIO()
for header, sequence in iterutils.chopped(tokens, 2):
sequence = sequence.upper()
unexpected_letters = set(sequence) - set('ACGT')
if unexpected_letters:
raise NexusError('unexpected sequence character(s): %s' % list(unexpected_letters))
print >> alignment_out, '>%s' % header
print >> alignment_out, sequence
alignment_string = alignment_out.getvalue()
self.alignment = Fasta.Alignment(StringIO(alignment_string))