def extract_target(seed_ref, coordinate_ref):
""" Extract a portion of the seed that aligns with the coordinate reference.
:param seed_ref: seed reference (nucleotide sequence)
:param coordinate_ref: coordinate reference (amino acid sequence)
:return: subsequence of seed_ref that maps to coordinate_ref
"""
best_alignment = (-1000000, '', '', 0)
for frame_index in range(3):
seed_aminos = translate('-' * frame_index + seed_ref)
aseed, acoord, score = align_it_aa(seed_aminos, coordinate_ref,
GAP_OPEN_COST, GAP_EXTEND_COST,
USE_TERMINAL_COST)
best_alignment = max(best_alignment,
(score, aseed, acoord, frame_index))
score, aseed, acoord, frame_index = best_alignment
assert score >= len(coordinate_ref) // 2, score
target = []
seed_index = -frame_index
for s, c in zip(aseed, acoord):
if s == '-':
continue
seed_index += 3
if c == '-':
continue
target.append(seed_ref[seed_index - 3:seed_index])
return ''.join(target)
def find_coord_pos(projects, coord_name, start_pos, end_pos):
coord_seq = projects.getReference(coord_name)
gap_open = 40
gap_extend = 10
use_terminal_gap_penalty = 1
highest_score = 0
best_match = None
for ref_name in sorted(projects.getProjectSeeds('HCV')):
if not ref_name.startswith('HCV-2'):
continue
ref_nuc_seq = projects.getReference(ref_name)
for nuc_offset in range(3):
ref_amino_seq = translate(ref_nuc_seq, nuc_offset)
aligned_coord, aligned_ref, score = align_it_aa(
coord_seq, ref_amino_seq, gap_open, gap_extend,
use_terminal_gap_penalty)
if score > highest_score:
highest_score = score
best_match = (ref_name, nuc_offset, aligned_coord, aligned_ref)
ref_name, nuc_offset, aligned_coord, aligned_ref = best_match
coord_pos = ref_pos = 0
ref_start = ref_end = None
for coord_amino, ref_amino in zip(aligned_coord, aligned_ref):
if coord_amino != '-':
coord_pos += 1
if ref_amino != '-':
ref_pos += 1
if start_pos == coord_pos:
ref_start = ref_pos * 3 - nuc_offset - 3
if coord_pos == end_pos:
ref_end = ref_pos * 3 - nuc_offset
return ref_name, ref_start, ref_end
def extract_target(seed_ref, coordinate_ref):
""" Extract a portion of the seed that aligns with the coordinate reference.
:param seed_ref: seed reference (nucleotide sequence)
:param coordinate_ref: coordinate reference (amino acid sequence)
:return: subsequence of seed_ref that maps to coordinate_ref
"""
best_alignment = (-1000000, '', '', 0)
for frame_index in range(3):
seed_aminos = translate('-'*frame_index + seed_ref)
aseed, acoord, score = align_it_aa(seed_aminos,
coordinate_ref,
GAP_OPEN_COST,
GAP_EXTEND_COST,
USE_TERMINAL_COST)
best_alignment = max(best_alignment, (score, aseed, acoord, frame_index))
score, aseed, acoord, frame_index = best_alignment
assert score >= len(coordinate_ref) // 2, score
target = []
seed_index = -frame_index
for s, c in zip(aseed, acoord):
if s == '-':
continue
seed_index += 3
if c == '-':
continue
target.append(seed_ref[seed_index-3:seed_index])
return ''.join(target)
def _pair_align(self, reference, query, gap_open=15, gap_extend=5, use_terminal_gap_penalty=1):
""" Align a query sequence to a reference sequence.
@return: (aligned_ref, aligned_query, score)
"""
aligned_ref, aligned_query, score = gotoh.align_it_aa(
reference,
query,
gap_open,
gap_extend,
use_terminal_gap_penalty)
return aligned_ref, aligned_query, score
def _pair_align(self, reference, query, gap_open=15, gap_extend=5, use_terminal_gap_penalty=1):
""" Align a query sequence of amino acids to a reference sequence.
@return: (aligned_ref, aligned_query, score)
"""
# noinspection PyUnresolvedReferences
aligned_ref, aligned_query, score = gotoh.align_it_aa(
reference,
query,
gap_open,
gap_extend,
use_terminal_gap_penalty)
return aligned_ref, aligned_query, score
def _pair_align(self, reference, query, gap_open=15, gap_extend=5, use_terminal_gap_penalty=1):
""" Align a query sequence of amino acids to a reference sequence.
@return: (aligned_ref, aligned_query, score)
"""
# noinspection PyUnresolvedReferences
aligned_ref, aligned_query, score = gotoh.align_it_aa(
reference,
query,
gap_open,
gap_extend,
use_terminal_gap_penalty)
return aligned_ref, aligned_query, score
def align_it_aa(self,
seqa,
seqb,
gap_ini,
gap_ext,
use_terminal_gap_penalty=False,
emulate_rb=False):
'''
Returns aligned sequences (with gaps) from the Gotoh algorithm.
Expects amino acid sequences, see align_it() for nucleotide.
Parameters:
seqa (string): Amino acid sequence (standard)
seqb (string): Another amino acid sequence (seq)
gap_init (int): Gap initialization penalty
gap_extend (int): Gap extension penalty
use_terminal_gap_penalty (bool): penalize trailing gaps?
emulate_rb (bool): use original (Ruby) match/mismatch scores?
Returns:
seqa (string): Aligned sequence a
seqb (string): Aligned sequence b
score (int): alignment score (gap penalties + match/mismatch)
exit_status (AlignItResult): ok, illegal_char, internal_error
'''
sa = ""
sb = ""
score = 0
al_status = AlignItResult.internal_error
try:
if not bool(
self.valid_nu.search(seqa) and self.valid_nu.search(seqb)):
al_status = AlignItResult.illegal_char
else:
if emulate_rb:
[sa, sb] = gotoh.align_it_aa_rb(seqa, seqb, gap_ini,
gap_ext)
score = 0
else:
[sa, sb,
score] = gotoh.align_it_aa(seqa, seqb, gap_ini, gap_ext,
int(use_terminal_gap_penalty))
al_status = AlignItResult.ok
except:
al_status = AlignItResult.internal_error
return sa, sb, score, al_status
def find_coord_pos(projects: ProjectConfig,
coord_name: str,
start_pos: int = None,
end_pos: int = None):
coord_seq = projects.getReference(coord_name)
if start_pos is None:
start_pos = 1
if end_pos is None:
end_pos = len(coord_seq) + 1
if projects.config['regions'][coord_name]['is_nucleotide']:
# Already have a nucleotide sequence, nothing to do.
return coord_name, start_pos, end_pos
gap_open = 40
gap_extend = 10
use_terminal_gap_penalty = 1
highest_score = 0
best_match = None
ref_names = set()
for project in projects.config['projects'].values():
for region in project['regions']:
if coord_name == region['coordinate_region']:
ref_names.update(region['seed_region_names'])
for ref_name in sorted(ref_names):
ref_nuc_seq = projects.getReference(ref_name)
for nuc_offset in range(3):
ref_amino_seq = translate(ref_nuc_seq, nuc_offset)
aligned_coord, aligned_ref, score = align_it_aa(
coord_seq, ref_amino_seq, gap_open, gap_extend,
use_terminal_gap_penalty)
if score > highest_score:
highest_score = score
best_match = (ref_name, nuc_offset, aligned_coord, aligned_ref)
ref_name, nuc_offset, aligned_coord, aligned_ref = best_match
coord_pos = ref_pos = 0
ref_start = ref_end = None
for coord_amino, ref_amino in zip(aligned_coord, aligned_ref):
if ref_amino != '-':
ref_pos += 1
if coord_amino != '-':
coord_pos += 1
if start_pos == coord_pos:
ref_start = ref_pos * 3 - nuc_offset - 3
if coord_pos == end_pos:
ref_end = ref_pos * 3 - nuc_offset
assert ref_start is not None
assert ref_end is not None
return ref_name, ref_start, ref_end
def find_coord_pos(projects, coord_name, start_pos, end_pos):
coord_seq = projects.getReference(coord_name)
gap_open = 40
gap_extend = 10
use_terminal_gap_penalty = 1
highest_score = 0
best_match = None
ref_names = set()
for project in projects.config['projects'].values():
for region in project['regions']:
if coord_name == region['coordinate_region']:
ref_names.update(region['seed_region_names'])
for ref_name in sorted(ref_names):
ref_nuc_seq = projects.getReference(ref_name)
for nuc_offset in range(3):
ref_amino_seq = translate(ref_nuc_seq, nuc_offset)
aligned_coord, aligned_ref, score = align_it_aa(
coord_seq,
ref_amino_seq,
gap_open,
gap_extend,
use_terminal_gap_penalty)
if score > highest_score:
highest_score = score
best_match = (ref_name, nuc_offset, aligned_coord, aligned_ref)
ref_name, nuc_offset, aligned_coord, aligned_ref = best_match
coord_pos = ref_pos = 0
ref_start = ref_end = None
for coord_amino, ref_amino in zip(aligned_coord, aligned_ref):
if ref_amino != '-':
ref_pos += 1
if coord_amino != '-':
coord_pos += 1
if start_pos == coord_pos:
ref_start = ref_pos * 3 - nuc_offset - 3
if coord_pos == end_pos:
ref_end = ref_pos * 3 - nuc_offset
assert ref_start is not None
assert ref_end is not None
return ref_name, ref_start, ref_end
def find_best_sequence(sequences):
best_match = (0, None, None) # (score, name, sequence)
for name, nuc_seq in sorted(sequences.items()):
nuc_seq = nuc_seq.replace('\n', '')
nuc_seq = nuc_seq.replace('*', '-').replace('?', '-')
for frame in range(3):
offset_seq = '-' * frame + nuc_seq
aa_seq = aln2counts.translate(offset_seq)
aseq, aref, score = align_it_aa(aa_seq, PSSM_V3LOOP, GAP_OPEN_COST,
GAP_EXTEND_COST, USE_TERMINAL_COST)
best_match = max(best_match, (score, name, aseq, aref))
if score >= 272:
pairs = zip(aseq, aref)
diffs = [' ' if a == b else '*' for a, b in pairs]
print('score', score, name)
print('result ', aseq)
print('diffs ',
''.join(diffs) if aseq != aref else 'no diffs')
print('compare', aref)
return best_match[:2]