def align_sequences(query, reference, o):
nuc_matrix = parasail.matrix_create("ACGT", 2, -1)
result_trace = parasail.nw_trace_striped_sat(query, reference, 3, 2,
nuc_matrix)
traceback = result_trace.get_traceback('|', '.', ' ')
print(traceback.ref)
print(traceback.comp)
print(traceback.query)
snps = get_snp_locs(traceback, o)
alignment_string = ""
for i in range(0, len(traceback.ref), 60):
alignment_string += f"Ref\t{i}\t{traceback.ref[i:i+60]}\n"
gap = ''
for j in range(len(str(i))):
gap += ' '
alignment_string += f" \t{gap}\t{traceback.comp[i:i+60]}\n"
alignment_string += f"CNS\t{i}\t{traceback.query[i:i+60]}\n"
alignment_string += "\n"
o.write("\n*****\n\n### Alignment with best reference\n\n")
o.write(alignment_string + "\n*****\n")
def generate_paf_file(query_fasta, target_fasta, output_file):
user_matrix = parasail.matrix_create("ACGT", 2, -2)
with open(target_fasta, 'r') as hin:
for line in hin:
if line.startswith('>'):
tid = line.rstrip('\n').split(' ')[0].lstrip('>')
else:
tseq = line.rstrip('\n')
with open(query_fasta, 'r') as hin, open(output_file, 'w') as hout:
for line in hin:
if line.startswith('>'):
qid = line.rstrip('\n').lstrip('>')
else:
qseq = line.rstrip('\n')
res = parasail.ssw(qseq, tseq, 3, 1, user_matrix)
if res is not None:
print("%s\t%d\t%d\t%d\t+\t%s\t%d\t%d\t%d\t*\t*\t60" %
(qid, len(qseq), res.read_begin1, res.read_end1, tid,
len(tseq), res.ref_begin1, res.ref_end1),
file=hout)
else:
logger.warning(
f'Error occured in the alignment of {qid} and {tid} via parasail'
)
def get_refined_bp_sbnd(tconsensus, fasta_file_h, tchr, tstart, tend, tdir, hout_log, margin = 200):
tconsensus_part = tconsensus[:1000] if len(tconsensus) > 1000 else tconsensus
ref_len = fasta_file_h.get_reference_length(tchr)
if tstart < 1: tstart = 1
if ref_len < tend: tend = ref_len
if tdir == '+':
qseq = fasta_file_h.fetch(tchr, max(int(tstart) - margin, 0), int(tend))
else:
qseq = fasta_file_h.fetch(tchr, max(int(tstart) - 1, 0), int(tend) + margin)
qseq = reverse_complement(qseq)
user_matrix = parasail.matrix_create("ACGT", 1, -2)
res = parasail.ssw(qseq, tconsensus, 3, 1, user_matrix)
if res is None:
logger.debug(f"Alignment for breakpoint localization failed for {tchr},{tstart},{tend},{tdir}")
return None
if tdir == '+':
bp_pos_reference = tend - (len(qseq) - res.read_end1 - 1)
else:
bp_pos_reference = tstart + (len(qseq) - res.read_end1 - 1)
tconsensus_after = tconsensus[(res.ref_end1 + 1):]
return (bp_pos_reference, tconsensus_after)
def generate_paf_file(self, query_fasta, target_fasta, output_file):
user_matrix = parasail.matrix_create("ACGT", 2, -2)
paf_rec_count = 0
with open(target_fasta, 'r') as hin:
for line in hin:
if line.startswith('>'):
tid = line.rstrip('\n').split(' ')[0].lstrip('>')
else:
tseq = line.rstrip('\n')
with open(query_fasta, 'r') as hin, open(output_file, 'w') as hout:
for line in hin:
if line.startswith('>'):
qid = line.rstrip('\n').lstrip('>')
else:
qseq = line.rstrip('\n')
res = parasail.ssw(qseq, tseq, 3, 1, user_matrix)
if res is not None:
print(f"{qid}\t{len(qseq)}\t{res.read_begin1}\t{res.read_end1}\t+\t" +
f"{tid}\t{len(tseq)}\t{res.ref_begin1}\t{res.ref_end1}\t*\t*\t60", file = hout)
paf_rec_count = paf_rec_count + 1
else:
self.parasail_error.append((qid, tid))
return(paf_rec_count)
def parasail_alignment(query, ref):
user_matrix = parasail.matrix_create("ACGT", 2, -20)
result = parasail.nw_trace_scan_16(query, ref, 50, 0, user_matrix)
if result.saturated:
print("SATURATED!")
else:
return cigar_to_seq(result.cigar.decode, query, ref)
def ssw_check_parasail(query, target):
user_matrix = parasail.matrix_create("ACGT", 2, -2)
alignment_info = {}
for sQId, sQSeq, sQQual in read(query):
sQSeq_r = reverse_complement(sQSeq)
for sTId, sTSeq, STQual in read(target):
res = parasail.ssw(sQSeq, sTSeq, 3, 1, user_matrix)
res_r = parasail.ssw(sQSeq_r, sTSeq, 3, 1, user_matrix)
if res.score1 > res_r.score1:
score = res.score1
qstart, qend = res.read_begin1 + 1., res.read_end1 + 1
tstart, tend = res.ref_begin1 + 1, res.ref_end1 + 1
strand = '+'
else:
score = res_r.score1
qstart, qend = len(sQSeq) - res_r.read_end1, len(
sQSeq) - res_r.read_begin1
tstart, tend = res_r.ref_begin1 + 1, res_r.ref_end1 + 1
strand = '-'
alignment_info[sTId] = [
score,
int(qstart),
int(qend),
int(tstart),
int(tend), strand
]
return (alignment_info)
def parasail_alignment(s1,
s2,
match_score=2,
mismatch_penalty=-2,
opening_penalty=3,
gap_ext=1):
user_matrix = parasail.matrix_create("ACGT", match_score, mismatch_penalty)
result = parasail.sg_trace_scan_16(s1, s2, opening_penalty, gap_ext,
user_matrix)
if result.saturated:
print("SATURATED!", len(s1), len(s2))
result = parasail.sg_trace_scan_32(s1, s2, opening_penalty, gap_ext,
user_matrix)
print("computed 32 bit instead")
# difference in how to obtain string from parasail between python v2 and v3...
if sys.version_info[0] < 3:
cigar_string = str(result.cigar.decode).decode('utf-8')
else:
cigar_string = str(result.cigar.decode, 'utf-8')
s1_alignment, s2_alignment, cigar_tuples = cigar_to_seq(
cigar_string, s1, s2)
# print(result.score, len(s1), len(s2))
# print(s1_alignment)
# print(s2_alignment)
# print(cigar_string)
# sys.exit()
# print(dir(result))
# print(result.end_query, result.end_ref, result.len_query, result.len_ref, result.length, result.matches)
# print()
return s1_alignment, s2_alignment, cigar_string, cigar_tuples, result.score
def align_secondary_reference(primary_flank, secondary_ref):
"""
An seqan alignment of a primary reference slice against a complete
secondary reference.
"""
MATRIX = parasail.matrix_create("ACGT", 2, -1)
OPEN = 2
EXTEND = 1
# Semi-Global, do not penalize gaps at beginning and end of s2/database
trace = parasail.sg_dx_trace_striped_sat(str(primary_flank.seq),
str(secondary_ref.seq), OPEN,
EXTEND, MATRIX)
traceback = trace.get_traceback()
query_end = trace.end_query
ref_end = trace.end_ref
aligned_query = traceback.query[ref_end - query_end:ref_end + 1]
aligned_ref = traceback.ref[ref_end - query_end:ref_end + 1]
# Alignment failed (indels)
if "-" in aligned_query + aligned_ref or len(primary_flank) != len(
aligned_ref):
raise FailedAlignmentError(
"Alignment failed between primary and secondary reference.",
reference=secondary_ref,
)
return SeqRecord(Seq(aligned_ref), id=secondary_ref.id)
def align_affine(query: str, target: str, match_score: int, mismatch_cost: int,
gap_open_cost: int, gap_extension_cost: int) -> Cigar:
assert min(match_score, mismatch_cost, gap_open_cost,
gap_extension_cost) >= 0, "Specify positive integers"
return Cigar(
parasail.nw_trace(
query, target, gap_open_cost, gap_extension_cost,
parasail.matrix_create(
"ACGT", match_score,
-mismatch_cost)).cigar.decode.decode('utf-8'))
def parasail_block_alignment(s1,
s2,
k,
match_id,
x_acc="",
y_acc="",
match_score=2,
mismatch_penalty=-2,
opening_penalty=5,
gap_ext=1,
ends_discrepancy_threshold=0):
user_matrix = parasail.matrix_create("ACGT", match_score, mismatch_penalty)
result = parasail.sg_trace_scan_16(s1, s2, opening_penalty, gap_ext,
user_matrix)
if result.saturated:
print("SATURATED!")
result = parasail.sg_trace_scan_32(s1, s2, opening_penalty, gap_ext,
user_matrix)
if sys.version_info[0] < 3:
cigar_string = str(result.cigar.decode).decode('utf-8')
else:
cigar_string = str(result.cigar.decode, 'utf-8')
s1_alignment, s2_alignment = cigar_to_seq(cigar_string, s1, s2)
# Rolling window of matching blocks
# k=15
# match_id = int(k*0.8) 1.0 - math.ceil(window_fraction)
match_vector = [
1 if n1 == n2 else 0 for n1, n2 in zip(s1_alignment, s2_alignment)
]
# print("".join([str(m) for m in match_vector]))
match_window = deque(match_vector[:k]) # initialization
current_match_count = sum(match_window)
aligned_region = []
if current_match_count >= match_id:
aligned_region.append(1)
else:
aligned_region.append(0)
for new_m_state in match_vector[k:]:
prev_m_state = match_window.popleft()
current_match_count = current_match_count - prev_m_state + new_m_state
match_window.append(new_m_state)
if current_match_count >= match_id:
aligned_region.append(1)
else:
aligned_region.append(0)
# print("".join([str(m) for m in aligned_region]))
# print("Aligned ratio (tot aligned/len(seq1):", sum(aligned_region)/float(len(s1)))
alignment_ratio = sum(aligned_region) / float(len(s1))
return (s1, s2, (s1_alignment, s2_alignment, alignment_ratio))
def parasail_alignment(query, ref):
user_matrix = parasail.matrix_create("ACGT", 2, -4)
result = parasail.nw_trace_scan_32(query, ref, 12, 1, user_matrix)
if result.saturated:
print("SATURATED!")
else:
if sys.version_info[0] < 3:
cigar_string = str(result.cigar.decode).decode('utf-8')
else:
cigar_string = str(result.cigar.decode, 'utf-8')
return cigar_string, cigar_to_seq(cigar_string, query, ref)
def __init__(self, elements, gap_open=3, gap_extend=3, matrix=parasail.blosum62, useIdentity=False):
self.sim_cache = {}
self.elements = elements
self.e2i = {e:i[0] for e,i in zip(elements, self.indices())}
self.i2e = {i[0]:e for e,i in zip(elements, self.indices())}
self.matrix = matrix
self.gap_extend = gap_extend
self.gap_open = gap_open
if useIdentity:
self.matrix = parasail.matrix_create(alphabet='ACDEFGHIKLMNPQRSTVWXY', match=1, mismatch=0)
self.paraParams = dict(open=self.gap_open, extend=self.gap_extend, matrix=self.matrix)
def parasail_block_alignment(s1,
s2,
k,
match_id,
match_score=2,
mismatch_penalty=-2,
opening_penalty=5,
gap_ext=1):
user_matrix = parasail.matrix_create("ACGT", match_score, mismatch_penalty)
result = parasail.sg_trace_scan_16(s1, s2, opening_penalty, gap_ext,
user_matrix)
if result.saturated:
print("SATURATED!", len(s1), len(s2))
result = parasail.sg_trace_scan_32(s1, s2, opening_penalty, gap_ext,
user_matrix)
print("computed 32 bit instead")
# difference in how to obtain string from parasail between python v2 and v3...
if sys.version_info[0] < 3:
cigar_string = str(result.cigar.decode).decode('utf-8')
else:
cigar_string = str(result.cigar.decode, 'utf-8')
s1_alignment, s2_alignment = help_functions.cigar_to_seq(
cigar_string, s1, s2)
# Rolling window of matching blocks
match_vector = [
1 if n1 == n2 else 0 for n1, n2 in zip(s1_alignment, s2_alignment)
]
match_window = deque(match_vector[:k]) # initialization
current_match_count = sum(match_window)
aligned_region = []
if current_match_count >= match_id:
aligned_region.append(1)
else:
aligned_region.append(0)
for new_m_state in match_vector[k:]:
prev_m_state = match_window.popleft()
current_match_count = current_match_count - prev_m_state + new_m_state
match_window.append(new_m_state)
if current_match_count >= match_id:
aligned_region.append(1)
else:
aligned_region.append(0)
# print("".join([str(m) for m in aligned_region]))
# print("Aligned ratio (tot aligned/len(seq1):", sum(aligned_region)/float(len(s1)))
alignment_ratio = sum(aligned_region) / float(len(s1))
return (s1, s2, (s1_alignment, s2_alignment, alignment_ratio))
def get_alignment(primer, reference):
"""An seqan alignment of a primer against a reference."""
MATRIX = parasail.matrix_create("ACGT", 2, -1)
OPEN = 2
EXTEND = 1
IDENTITY_THRESHOLD = 0.7
if primer.direction == Primer.Direction.left:
ref = reference.seq
elif primer.direction == Primer.Direction.right:
ref = reference.reverse_complement().seq
# Semi-Global, do not penalize gaps at beginning and end of s2/database
trace = parasail.sg_dx_trace_striped_sat(str(primer.seq), str(ref), OPEN,
EXTEND, MATRIX)
traceback = trace.get_traceback()
query_end = trace.end_query + 1
ref_end = trace.end_ref + 1
# Get alignment strings
aln_query = traceback.query[ref_end - query_end:ref_end]
cigar = traceback.comp[ref_end - query_end:ref_end]
aln_ref = traceback.ref[ref_end - query_end:ref_end]
# Identity for glocal alignment
identity = cigar.count("|") / len(cigar)
# Alignment failed
if identity < IDENTITY_THRESHOLD:
return None
# Format alignment
refid = reference.id[:30]
name = primer.name[:30]
formatted_query = f"{name: <30} {1: >6} {aln_query} {query_end}"
if primer.direction == Primer.Direction.left:
formatted_ref = (
f"{refid: <30} {ref_end - query_end + 1: >6} {aln_ref} {ref_end}")
elif primer.direction == Primer.Direction.right:
rev_start = len(reference) - ref_end
formatted_ref = (
f"{refid: <30} {rev_start + query_end: >6} {aln_ref} {rev_start + 1}"
)
formatted_cigar = f"{'': <30} {'': >6} {cigar}"
formatted_alignment = "\n".join(
["", formatted_query, formatted_cigar, formatted_ref])
del trace
return (identity, formatted_alignment)
def make_scoring_matrix(match_reward_coding):
match_reward_non_coding = 2
mismatch_penalty = -4
matrix = parasail.matrix_create("ACGTacgt*", match_reward_non_coding, mismatch_penalty, True)
matrix[0,4] = match_reward_coding
matrix[1,5] = match_reward_coding
matrix[2,6] = match_reward_coding
matrix[3,7] = match_reward_coding
matrix[4,0] = match_reward_coding
matrix[5,1] = match_reward_coding
matrix[6,2] = match_reward_coding
matrix[7,3] = match_reward_coding
return matrix
def test1():
matrix = parasail.blosum62
matrix[3, 4] = 100
print(matrix.matrix)
matrix = parasail.blosum62.copy()
matrix[3, 4] = 100
print(matrix.matrix)
matrix = parasail.matrix_create("ACGT", 10, 1)
matrix[2] = 200
matrix[1:3, 1:3] = 300
matrix.set_value(0, 4, 400)
print(matrix.matrix)
matrix = parasail.matrix_create("AcgT", 10, 1)
print(matrix.matrix)
print(matrix.mapper)
matrix = parasail.matrix_create("AcgT", 10, 1, True)
print(matrix.matrix)
print(matrix.mapper)
def main():
'''
Entry point for Readucks. Gets arguments, processes them and then calls process_files function
to do the actual work.
:return:
'''
args = get_arguments()
barcode_set = 'native'
# if args.native_barcodes:
# barcode_set = 'native'
if args.pcr_barcodes:
barcode_set = 'pcr'
if args.rapid_barcodes:
barcode_set = 'rapid'
settings = {
'barcode_set': "native",
'single_barcode': args.single,
'threshold': args.threshold / 100.0,
'secondary_threshold': None
}
if args.secondary_threshold:
settings['secondary_threshold'] = args.secondary_threshold / 100.0
# set_alignment_settings( 10,
# 1,
# parasail.matrix_create("ACGT", 3, -2))
set_alignment_settings(
-args.scoring_scheme_vals[2], -args.scoring_scheme_vals[3],
parasail.matrix_create("ACGT", args.scoring_scheme_vals[0],
args.scoring_scheme_vals[1]))
output_path = None
if args.output_dir:
if os.path.isdir(args.output_dir):
output_path = args.output_dir.rstrip("/") + "/"
output = {
'path': output_path,
'prefix': args.prefix,
'bin_barcodes': args.bin_barcodes,
'annotate_files': args.annotate_files,
'extended_info': args.extended_info,
'bin_files': {}
}
process_files(args.input_path, output, barcode_set, args.limit_barcodes_to,
settings, args.verbosity, args.threads)
def ssw_alignment(s1,
s2,
match_score=2,
mismatch_penalty=-2,
opening_penalty=3,
gap_ext=1):
user_matrix = parasail.matrix_create("ACGT", match_score, mismatch_penalty)
result = parasail.ssw(s1, s2, opening_penalty, gap_ext, user_matrix)
print(result, type(result), dir(result))
print(dir(result))
for attr, value in result.__dict__.items():
print(attr, value)
# print(result.ref_begin1, result.ref_end1, result.read_begin1, result.read_end1)
# print()
return s1_alignment, s2_alignment, cigar_string, cigar_tuples, result.score
def validate_reconstructed_seq(seq, orig):
"""
seq --- the sequence that is reconstructed
orig --- the original sequence
because the reconstructed seq can be longer, we don't care about deletions
(deletions w.r.t could just be exon skipping or minor base errors)
we only care that there is NOT a lot of insertions (which would indicate error in my bubble solution)
"""
o1 = parasail.sg_qx_trace(seq, orig, 3, 1, parasail.matrix_create("ACGT", 2, -5))
if o1.score < l2*2*.90: return False, o1.cigar.decode
for num, type in iter_cigar_string(o1.cigar.decode):
if type == 'I' and num > 5:
return False, o1.cigar.decode
return True, o1.cigar.decode
def pair_align(reference, query, params=DEFAULT_ALIGN_PARAMS):
""" Perform pairwise local alignment using scikit-bio.
:param reference: Reference sequence.
:param query: Query sequence.
:param params: Alignment parameters in a dictionary.
:returns: Alignments in scikit-bio format.
:rtype: list of tuples
"""
subs_mat = parasail.matrix_create("ACGT", params['match'],
params['mismatch'])
aln = parasail.sw_striped_32(reference, query, params['gap_open'],
params['gap_extend'], subs_mat)
return aln
def update_matrix(self):
"""
Create new parasail scoring matrix. 'N' is used as wildcard character
for barcodes and has its own match parameter (0 per default).
'X' is used as wildcard character for modified bp as in the 16S
sequencing adapter.
:return: None
"""
self.matrix = parasail.matrix_create("ATGCNX", self.match,
self.mismatch)
pointers = [4, 11, 18, 25, 28, 29, 30, 31, 32]
for i in pointers:
self.matrix.pointer[0].matrix[i] = self.nmatch
pointers = [5, 12, 19, 26, 33, 35, 36, 37, 38, 39, 40]
for i in pointers:
self.matrix.pointer[0].matrix[i] = 0
def create_matrix(match, mismatch, nmatch):
"""
Create new parasail scoring matrix. 'N' is used as wildcard character
for barcodes and has its own match parameter (0 per default).
'X' is used as wildcard character for modified bp as in the 16S
sequencing adapter. Taken from qcat.
:return: parasail matrix
"""
matrix = parasail.matrix_create("ATGCNX", match, mismatch)
pointers = [4, 11, 18, 25, 28, 29, 30, 31, 32]
for i in pointers:
matrix.pointer[0].matrix[i] = nmatch
pointers = [5, 12, 19, 26, 33, 35, 36, 37, 38, 39, 40]
for i in pointers:
matrix.pointer[0].matrix[i] = 0
return matrix
def __init__(self,
elements,
gap_open=3,
gap_extend=3,
matrix=parasail.blosum62,
useIdentity=False):
self.sim_cache = {}
self.elements = elements
self.e2i = {e: i[0] for e, i in zip(elements, self.indices())}
self.i2e = {i[0]: e for e, i in zip(elements, self.indices())}
self.matrix = matrix
self.gap_extend = gap_extend
self.gap_open = gap_open
if useIdentity:
self.matrix = parasail.matrix_create(
alphabet='ACDEFGHIKLMNPQRSTVWXY', match=1, mismatch=0)
self.paraParams = dict(open=self.gap_open,
extend=self.gap_extend,
matrix=self.matrix)
def node_is_similar(seq1, seq2):
l1 = len(seq1)
l2 = len(seq2)
if l1 == 0 or l2 == 0: return False
if l1 <= 2 and l2 <= 2: return True
if l1 < l2:
l1, l2 = l2, l1
seq1, seq2 = seq2, seq1
# always make seq1 the longer one
o1 = parasail.sg_qx_trace(seq1, seq2, 3, 1, parasail.matrix_create("ACGT", 2, -5))
# require the the whole (shorter) seq2 must be aligned
# and set min score to approx 90% accuracy
if EXPECTED_ERR_RATE == 0:
return o1.score > l2*2*1.0
elif EXPECTED_ERR_RATE < 2:
return o1.score > l1*2*0.8
else:
raise Exception("Expected error rate not implemented for {0}% and above".format(EXPECTED_ERR_RATE))
return res is not None
def __init__(self, config_path=None):
self._match = 5
self._nmatch = -1
self._mismatch = -2
self._gap_open = 2
self._gap_extend = 2
self._max_align_length = 150
self._extracted_barcode_extension = 11
self._barcode_context_length = 11
self._matrix = None
self.update_matrix()
self._matrix_barcode = parasail.matrix_create("ATGCN", 1, -1)
if config_path is not None:
self.read(config_path)
def parasail_local(s1,
s2,
match_score=2,
mismatch_penalty=-2,
opening_penalty=3,
gap_ext=1):
user_matrix = parasail.matrix_create("ACGT", match_score, mismatch_penalty)
result = parasail.sw_trace_scan_16(s1, s2, opening_penalty, gap_ext,
user_matrix)
if result.saturated:
print("SATURATED!", len(s1), len(s2))
result = parasail.sg_trace_scan_32(s1, s2, opening_penalty, gap_ext,
user_matrix)
print("computed 32 bit instead")
# difference in how to obtain string from parasail between python v2 and v3...
if sys.version_info[0] < 3:
cigar_string = str(result.cigar.decode).decode('utf-8')
else:
cigar_string = str(result.cigar.decode, 'utf-8')
s1_alignment, s2_alignment, cigar_tuples = cigar_to_seq(
cigar_string, s1[result.cigar.beg_query:result.end_query],
s2[result.cigar.beg_ref:result.end_ref])
# print(result.traceback.ref)
# print(result.traceback.comp)
# print(result.traceback.query)
# print(result.score, len(s1), len(s2))
print("read", s1_alignment)
print("Rref", s2_alignment)
print(result.cigar.beg_query, result.end_query)
print(result.cigar.beg_ref, result.end_ref)
print(cigar_string)
# print(result.cigar.seq)
# sys.exit()
# print(dir(result))
# for attr, value in result.__dict__.items():
# print(attr, value)
# print(result.end_query, result.end_ref, result.len_query, result.len_ref, result.length, result.matches)
# print()
return s1_alignment, s2_alignment, cigar_string, cigar_tuples, result.score
def parasail_alignment(read,
reference,
x_acc="",
y_acc="",
match_score=2,
mismatch_penalty=-2,
opening_penalty=2,
gap_ext=1,
ends_discrepancy_threshold=0):
user_matrix = parasail.matrix_create("ACGT", match_score, mismatch_penalty)
result = parasail.sg_trace_scan_16(read, reference, opening_penalty,
gap_ext, user_matrix)
if result.saturated:
print("SATURATED!")
result = parasail.sg_trace_scan_32(read, reference, opening_penalty,
gap_ext, user_matrix)
if sys.version_info[0] < 3:
cigar_string = str(result.cigar.decode).decode('utf-8')
else:
cigar_string = str(result.cigar.decode, 'utf-8')
read_alignment, ref_alignment = cigar_to_seq(cigar_string, read, reference)
return read_alignment, ref_alignment
def generate_paf_file(query_fasta, target_fasta, output_file):
user_matrix = parasail.matrix_create("ACGT", 2, -2)
with open(target_fasta, 'r') as hin:
for line in hin:
if line.startswith('>'):
tid = line.rstrip('\n').lstrip('>')
else:
tseq = line.rstrip('\n')
with open(query_fasta, 'r') as hin, open(output_file, 'w') as hout:
for line in hin:
if line.startswith('>'):
qid = line.rstrip('\n').lstrip('>')
else:
qseq = line.rstrip('\n')
res = parasail.ssw(qseq, tseq, 3, 1, user_matrix)
print("%s\t%d\t%d\t%d\t+\t%s\t%d\t%d\t%d\t*\t*\t60" %
(qid, len(qseq), res.read_begin1, res.read_end1, tid,
len(tseq), res.ref_begin1, res.ref_end1),
file=hout)
"""
def create_score_matrix(self):
"""Creates a parasail score matrix for alignment
"""
return parasail.matrix_create("ACGTN", self.score, -1 * self.penalty)
import os
import logging
import math
import dnaio
import gc
import parasail
from .fastq_file import ReadIdentifier
logger = logging.getLogger(__name__)
match_score = 0
DEFAULT_SCORE_MATRIX = parasail.matrix_create("ACGTN", match_score, -1)
class ReadPair:
def __init__(self, read1, read2):
self.read1 = read1
self.read2 = read2
id_1 = ReadIdentifier(self.read1.name)
id_2 = ReadIdentifier(self.read2.name)
if id_1.identifier != id_2.identifier:
raise ValueError(
"Identifiers of the read pairs does not match each other.\n"
"Read1: %s\n"
"Read2: %s" % (id_1.identifier, id_2.identifier))
self._identifier = id_1.identifier
if id_1.pair_member and id_2.pair_member:
if id_1.pair_member == id_2.pair_member:
def test3():
matrix = parasail.matrix_create("acgt", 1, -1)
result = parasail.sw("acgt", "acgt", 10, 1, matrix)
assert (result.score == 4)
del result
del matrix
def test21():
matrix = parasail.matrix_create("ACGTacgt", 2, 1, True)
result = parasail.sw_trace("ACGT","AcgT",10,1,matrix)
traceback = result.traceback
print_traceback_attributes(traceback)
def test4():
parasail.set_case_sensitive(True)
matrix = parasail.matrix_create("ACGT", 2, 1)
result = parasail.sw_trace("ACGT","AcgT",10,1,matrix)
traceback = result.traceback
print_traceback_attributes(traceback)
def test22():
matrix = parasail.matrix_create("ACGTacgt", 2, 1, True)
result = parasail.sw_trace("ACGT","AcgT",10,1,matrix)
traceback = result.get_traceback(case_sensitive=True)
print_traceback_attributes(traceback)