def __getitem__(self, args):
"""
Allows to get AlignmentPositions using an index
or a list of AlignmentPosition instances.
"""
#TODO: replace by list slice in alignment
if type(args) == int:
if args in range(1, len(self.alignment) + 1):
return self.alignment[args - 1]
else:
raise AlignmentError(
"There is no such position in the alignment: %i" % args)
elif type(args) == slice:
counter = args.start
returned_positions = {}
while counter <= args.stop:
if counter <= len(self.alignment):
returned_positions[counter] = self.alignment[counter - 1]
counter += 1
else:
raise AlignmentError("The alignment is shorter than %i" %
counter)
return returned_positions
else:
raise AlignmentError('Bad argument type.')
def get_alignment_from_file(self, filename, shrink=DEFAULT_SHRINK):
"""Returns a RNAAlignment object."""
try:
data = open(filename).read()
except IOError:
raise AlignmentError('File does not exist: %s ' % filename)
return self.get_alignment(data)
def get_differences(self, mode='is_different'):
"""
Returns these positions in alignment which are different
in a given sequences as a list of AlignmentPosition instances.
Arguments:
- mode (by default 'all')
Modes:
- 'is_different' - all differences are taken into account
('mismatch','gap in template','gap in target')
- 'has_gap' - only gaps in template and gaps in target
- 'has_template_gap'
- 'has_target_gap'
- 'is_mismatch'
- 'is_unidentified' - when one of letters or both is ANY_RESIDUE
"""
differences = []
if mode in MODES:
for apos in self.alignment:
method = getattr(apos, mode)
if method():
differences.append(apos)
return differences
raise AlignmentError("Bad mode type. Should be one of %s." %
str(MODES))
def set_aligned_sequences(self, char_tuples):
"""Resets the sequences in the RNAAlignment object."""
transposed = map(list, zip(*char_tuples))
target = Sequence(transposed[0])
template = Sequence(transposed[1])
if len(target) != len(template):
raise AlignmentError(
"Error correcting alignment; lenghts differ:\n%s\%s" %
(str(target), str(template)))
self.align.set_aligned_sequences(target, template)
def read_alignment(data, shrink=DEFAULT_SHRINK):
parser = RNAAlignmentParser()
if os.access(data, os.F_OK):
alignment = parser.get_alignment_from_file(data, shrink)
elif data.startswith('>'):
alignment = parser.get_alignment(data, shrink)
else:
raise AlignmentError(
'Alignment not in FASTA format or file does not exist: %s' % data)
log.write_message('Alignment loaded from %s:%s' % (data, str(alignment)))
return alignment
def set_aligned_sequences(self, seq1, seq2):
"""Sets alignment to a list of AlignmentPosition instances."""
self.alignment = []
self.aligned_sequences = (seq1, seq2)
i_alignment = 1
i_template = 1
i_target = 1
ali_length = len(seq1)
seq1 = seq1.seq_alphabet_list
seq2 = seq2.seq_alphabet_list
if not len(seq1) == len(seq2):
raise AlignmentError(
'Sequence lengths in alignment do not match (%i, %i)' %
(len(seq1), len(seq2)))
for pos in range(ali_length):
# default values for creating AlignmentPositions
target_letter = seq1[pos]
template_letter = seq2[pos]
target_position = i_target
template_position = i_template
alignment_position = i_alignment
if seq1[pos].short_abbrev in ['-', '_'] \
and seq2[pos].short_abbrev in ['-', '_']:
continue
elif seq2[pos].short_abbrev == '-':
template_position = None
template_letter = None
i_target += 1
elif seq1[pos].short_abbrev == '-':
target_position = None
target_letter = None
i_template += 1
else:
i_template += 1
i_target += 1
i_alignment += 1
apos = AlignmentPosition(target_position=target_position,
target_letter=target_letter,
template_position=template_position,
template_letter=template_letter,
alignment_position=alignment_position)
self.alignment.append(apos)
def get_alignment(self, fasta_string, shrink=DEFAULT_SHRINK):
"""Returns a RNAAlignment object."""
temp_aln = []
for hit in fasta_string.split('\n>'):
defline = hit.split('\n', 1)[0].replace('>', '')
sequence = hit.split('\n', 1)[1:]
sequence = re.sub("[\r\s\t]+", '', ''.join(sequence))
temp_aln.append((defline, Sequence(sequence)))
name1, seq1 = temp_aln[0]
name2, seq2 = temp_aln[1]
alignment = RNAAlignment(name1, seq1, name2, seq2, shrink)
#TODO: remove this check after kicking underscores out.
for apos in alignment:
if not apos.check_underscore():
raise AlignmentError(
'Underscore symbols must be aligned to other underscores or gaps.'
)
return alignment