def pairwise_align_dna(sequence, reference, regex_complied, gene):
"""
Pairwise align sequence to reference, to find reading frame and frame-shift in/dels
:param sequence: (str) a query DNA sequence
:param reference: (str) a reference DNA sequence (must start in reading frame 1)
:param regex_complied: (regex_obj) a compiled regex pattern
:param gene: (str) the target gene (ENV, GAG, POL, etc...
:return: (str) aligned query sequence, (str) aligned ref sequence, (int) reading frame for query sequence
"""
# do overlap pairwise alignment to not get truncated query sequence
if gene == "ENV":
overlap = seqanpy.align_overlap(sequence, reference, band=-1, score_match=4, score_mismatch=-1, score_gapext=-3,
score_gapopen=-14)
else:
# for other regions
overlap = seqanpy.align_overlap(sequence, reference, band=-1, score_match=4, score_mismatch=-2, score_gapext=-3,
score_gapopen=-14)
overlap = list(overlap)
seq_align = overlap[1]
ref_align = overlap[2]
# print(">sqseq1\n{}\n".format(seq_align))
# print(">sqref1\n{}\n".format(ref_align))
# get start position in the seq, if not starting at index 0
if seq_align[0] == '-':
seq_start = regex_complied.search(seq_align).end()
else:
seq_start = 0
# get end position in the seq, if not starting at index 0
if seq_align[-1] == '-':
# reverse the string and find first non-gap character, mult match.end by -1 to get non-reversed index
seq_end = (regex_complied.search(seq_align[::-1]).end()) * -1
else:
seq_end = None
# ref start will be 0 for align_overlap
if ref_align[0] == '-':
ref_start = regex_complied.search(ref_align).end()
else:
ref_start = 0
# calculate reading frame (reference must start in frame 0)
frame = (seq_start - ref_start) % 3
# truncate the overlap alignment to the region of interest
seq_align = seq_align[seq_start:seq_end]
ref_align = ref_align[seq_start:seq_end]
# print(">sqseq2\n{}\n".format(seq_align))
# print(">sqref2\n{}\n".format(ref_align))
return seq_align, ref_align, frame
def get_consensus(self, region, PCR=1):
'''Get consensus for this sample'''
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
from Bio import SeqIO
from seqanpy import align_overlap
(fragment, start,
stop) = self.get_fragmented_roi(region,
VERBOSE=0,
include_genomewide=True)
seq = SeqIO.read(self.get_consensus_filename(fragment, PCR=PCR),
'fasta')
refseq = SeqIO.read(
self.get_reference_filename(fragment, format='fasta'),
'fasta')[start:stop]
score, ali1, ali2 = align_overlap(seq, refseq)
start = len(ali2) - len(ali2.lstrip('-'))
end = len(ali2.rstrip('-'))
seq_region = ali1[start:end]
seq = SeqRecord(Seq(seq_region, seq.seq.alphabet),
id=self.name,
name=self.name,
description=self.name)
return seq
def merge_allele_counts(ref_genomewide, acs, VERBOSE=0):
'''Merge the allele counts of all fragments
Note: we do not require full coverage of all fragments, the missing
ones will just have zero counts. Sometimes, cherry-picking the data
fragment by fragment might be a better choice.
'''
from hivwholeseq.utils.miseq import alpha, read_types
from seqanpy import align_overlap
ac = np.zeros((len(read_types), len(alpha), len(ref_genomewide)), int)
pos_ref = 1000
for (fr, ref, acsi) in acs:
# Find the coordinates
(score, ali1, ali2) = align_overlap(ref_genomewide[pos_ref - 1000:],
ref,
#score_gapopen=-20,
)
fr_start = len(ali2) - len(ali2.lstrip('-'))
fr_end = len(ali2.rstrip('-'))
if VERBOSE:
print fr, pos_ref - 1000 + fr_start, pos_ref - 1000 + fr_end
# Scan the alignment
pos_ref = pos_ref - 1000 + fr_start
fr_start_ref = pos_ref
fr_end_ref = pos_ref + fr_end - fr_start
pos_fr = 0
for pos_ali in xrange(fr_start, fr_end):
# Gap in genomewise, ignore position
if ali1[pos_ali] == '-':
pos_fr += 1
continue
# Gap in fragment, ignore FIXME: probably we should put deletions
elif ali2[pos_ali] == '-':
pos_ref += 1
continue
# Add the counts
# NOTE: all fragments are treated the same, even in case of coverage
# differences of orders of magnitude. This means, larger coverage
# always wins. Maybe we want to implement this somewhat differently
ac[:, :, pos_ref] += acsi[:, :, pos_fr]
pos_fr += 1
pos_ref += 1
if VERBOSE >= 3:
from hivwholeseq.utils.sequence import pretty_print_pairwise_ali
cons = alpha[ac.sum(axis=0).argmax(axis=0)]
pretty_print_pairwise_ali((ali1[fr_start: fr_end],
cons[fr_start: fr_end]),
name1='gw',
name2=fr,
width=100)
return ac
def get_distance_reads_sequence(seq,
reads,
VERBOSE=0,
score_match=3,
score_mismatch=-3):
'''Get the distance in alignment score between read pairs and a sequence'''
from seqanpy import align_overlap
seqs = ''.join(seq)
deltas = []
for irp, read_pair in enumerate(reads):
d = 0
for read in read_pair:
(score, alis, alir) = align_overlap(seqs,
read.seq,
score_match=score_match,
score_mismatch=score_mismatch)
start = len(alir) - len(alir.lstrip('-'))
end = len(alir.rstrip('-'))
scoremax = score_match * (end - start)
delta = scoremax - score
d += delta
deltas.append(d)
return deltas
def check_suspect(reads, consensi_foreign, deltamax=30, VERBOSE=0):
'''Check suspicious reads for closer distance to potential contaminants'''
if VERBOSE >= 2:
print 'Checking suspect read pair:', reads[0].qname,
for consensus in consensi_foreign:
conss = ''.join(consensus)
delta_foreign = 0
for read in reads:
(score, ali1, ali2) = align_overlap(conss, read.seq)
scoremax = 3 * len(ali2.strip('-'))
delta_foreign += scoremax - score
# We classify as trash all reads that are within a basin of another seq
if delta_foreign < deltamax:
if VERBOSE >= 2:
print ''
print consensus.name, delta_foreign
return True
else:
if VERBOSE >= 2:
print 'OK',
if VERBOSE >= 2:
print ''
return False
def check_suspect(reads, consensi_foreign, deltamax=30, VERBOSE=0):
'''Check suspicious reads for closer distance to potential contaminants'''
if VERBOSE >= 2:
print 'Checking suspect read pair:', reads[0].qname,
for consensus in consensi_foreign:
conss = ''.join(consensus)
delta_foreign = 0
for read in reads:
(score, ali1, ali2) = align_overlap(conss, read.seq)
scoremax = 3 * len(ali2.strip('-'))
delta_foreign += scoremax - score
# We classify as trash all reads that are within a basin of another seq
if delta_foreign < deltamax:
if VERBOSE >= 2:
print ''
print consensus.name, delta_foreign
return True
else:
if VERBOSE >= 2:
print 'OK',
if VERBOSE >= 2:
print ''
return False
def ammend_fasta():
with open(fname_unspliced, 'w') as ofile:
strain_by_protein = defaultdict(dict)
for seq in SeqIO.parse(fname, 'fasta'):
SeqIO.write(seq, ofile, 'fasta')
prot = seq.name.split('_')[0]
seq_name = seq.description.split('|')[1]
strain_by_protein[prot][seq_name] = seq
joined_seqs = defaultdict(dict)
splice_pairs = [('M', 'M1', 'M2'), ('NS', 'NS1', 'NS2')]
#splice_pairs = [('M', 'M1', 'BM2')]
from seqanpy import align_overlap
for c, a,b in splice_pairs:
for strain in strain_by_protein[a]:
seq1 = strain_by_protein[a][strain]
new_id = c+seq1.id.lstrip(a)
new_description = c+seq1.description.lstrip(a)
new_description = new_description.replace(a, c)
new_name = c+seq1.name.lstrip(a)
new_name = new_name.replace(a, c)
try:
score, ali1, ali2 = align_overlap(seq1.seq, strain_by_protein[b][strain].seq,
score_gapopen=-20, score_gapext=0)
ali_array = np.array([np.fromstring(x, 'S1') for x in [ali1, ali2]])
tmpseq = np.copy(ali_array[0])
tmpseq[ali_array[0]=='-']=ali_array[1][ali_array[0]=='-']
joined_seqs[c][strain] = SeqRecord.SeqRecord(seq=Seq.Seq("".join(tmpseq)), id=new_id,
description=new_description, name=new_name)
SeqIO.write(joined_seqs[c][strain], ofile, 'fasta')
except:
print(seq1.name, "doesn't have a partner")
def determine_lineage(seq):
fields = map(lambda x:x.strip(), seq.description.split('|'))
tmp_lineage = (fields[2], fields[4])
if tmp_lineage in patterns:
print fields[0],"\n\tgisaid defined lineage:",tmp_lineage,'->',patterns[tmp_lineage]
return patterns[tmp_lineage]
else:
scores = []
for olineage, oseq in outgroups.iteritems():
if (params.aligner == "seqan"):
from seqanpy import align_overlap
tmp_aln = align_overlap(str(oseq.seq), str(seq.seq).replace('-','').upper(),
score_gapopen=-10, score_gapext=-1)
tmp_aln = np.array([np.fromstring(tmp_aln[1], dtype='|S1'), np.fromstring(tmp_aln[2], dtype='|S1')])
if (params.aligner == "mafft"):
SeqIO.write([oseq, seq], "temp_in.fasta", "fasta")
os.system("mafft --auto temp_in.fasta > temp_out.fasta 2>tmp")
tmp_aln = np.array(AlignIO.read('temp_out.fasta', 'fasta'))
scores.append((olineage, (tmp_aln[0]==tmp_aln[1]).sum()))
scores.sort(key = lambda x:x[1], reverse=True)
if scores[0][1]>0.85*len(seq):
print fields[0], tmp_lineage, len(seq), "\n\t lineage based on similarity:",scores[0][0],"\n\t",scores
return scores[0][0]
else:
print fields[0], tmp_lineage, len(seq), "\n\t other: best scores:",scores[0]
return 'other'
def align_to_initial(reads, p):
from hivevo.sequence import alphal
seg, start, stop = 'F1', int(p.annotation['p17'].location.start)-20, int(p.annotation['p17'].location.end)+7
ref_seq = "".join(p.get_initial_sequence(seg)[start:stop])
aft = p.get_allele_frequency_trajectories('F1')[:,:,start:stop]
founder_indices = np.array([alphal.index(nuc) for nuc in ref_seq])
for read in reads:
score, ali_ref, ali_read = align_overlap(ref_seq, str(read.seq.ungap('-')))
tmp_ali = np.vstack([np.fromstring(a, 'S1') for a in (ali_ref, ali_read)])
try:
unconserved = ~((aft[:,:4,:].max(axis=1)>0.99).all(axis=0))
reference_aln = (tmp_ali[0]!='-')
ungapped = (~np.any(tmp_ali=='-', axis=0))[reference_aln]
unamb = (~np.any(tmp_ali=='N', axis=0))[reference_aln]
good_positions = ungapped&unamb&unconserved
read_indices = np.array([alphal.index(nuc) for nuc in tmp_ali[1]])[reference_aln][good_positions]
read.prob = np.sum(np.log(aft[:, read_indices, good_positions] + 0.001), axis=1)
good_positions = ungapped&unamb
good_positions[:20]=False
good_positions[-7:]=False
read_indices = np.array([alphal.index(nuc) for nuc in tmp_ali[1]])[reference_aln][good_positions]
read.distance = np.mean((founder_indices[good_positions]!=read_indices))
except:
import ipdb; ipdb.set_trace();
def determine_lineage(seq):
fields = map(lambda x:x.strip(), seq.description.split('|'))
tmp_lineage = (fields[2], fields[4])
if tmp_lineage in patterns:
# print fields[0],"\n\tgisaid defined lineage:",tmp_lineage,'->',patterns[tmp_lineage]
return patterns[tmp_lineage]
else:
scores = []
for olineage, oseq in outgroups.iteritems():
if (params.aligner == "seqan"):
from seqanpy import align_overlap
tmp_aln = align_overlap(str(oseq.seq), str(seq.seq).replace('-','').upper(),
score_gapopen=-10, score_gapext=-1)
tmp_aln = np.array([np.fromstring(tmp_aln[1], dtype='|S1'), np.fromstring(tmp_aln[2], dtype='|S1')])
if (params.aligner == "mafft"):
SeqIO.write([oseq, seq], "temp_in.fasta", "fasta")
os.system("mafft --auto temp_in.fasta > temp_out.fasta 2>tmp")
tmp_aln = np.array(AlignIO.read('temp_out.fasta', 'fasta'))
scores.append((olineage, (tmp_aln[0]==tmp_aln[1]).sum()))
scores.sort(key = lambda x:x[1], reverse=True)
if scores[0][1]>0.85*len(seq):
print fields[0], tmp_lineage, len(seq), "\n\t lineage based on similarity:",scores[0][0],"\n\t",scores
return scores[0][0]
else:
print fields[0], tmp_lineage, len(seq), "\n\t other: best scores:",scores[0]
return 'other'
def get_gene_positions_in_fragment(gene, fragment, gwseq, fragseq, VERBOSE=0):
'''Get the coordinates of a gene within a fragment'''
# Find coordinates of gene in reference
feagene = gwseq.features[map(attrgetter('id'), gwseq.features).index(gene)]
gene_start = feagene.location.nofuzzy_start
gene_end = feagene.location.nofuzzy_end
# Sanity check on coordinates
feafrag = refseq.features[map(attrgetter('id'),
gwseq.features).index(fragment)]
fragrefgw = feafrag.extract(gwseq)
if len(fragseq) != len(fragrefgw):
raise ValueError(
'Problem with coordinates between fragment and genomewide.')
# Find coordinates of gene in fragment
frag_start = feafrag.location.nofuzzy_start
frag_end = feafrag.location.nofuzzy_end
# complete gene
if (frag_start <= gene_start) and (frag_end >= gene_end):
if VERBOSE >= 2:
print 'Complete gene found'
positions = np.arange(gene_start, gene_end) - frag_start
# start of gene
elif (frag_start <= gene_start):
if VERBOSE >= 2:
print 'WARNING: only gene start found'
positions = np.arange(gene_start, frag_end) - frag_start
if len(positions) % 3:
positions = positions[:-(len(positions) % 3)]
# end of gene
elif (frag_end >= gene_end):
if VERBOSE >= 2:
print 'WARNING: only gene end found'
positions = np.arange(frag_start, gene_end) - frag_start
if len(positions) % 3:
positions = positions[len(positions) % 3:]
# middle of gene: guess reading frame
else:
if VERBOSE >= 2:
print 'WARNING: only gene middle found'
prot = feagene.extract(gwseq).seq.translate()
ali_score = []
for rf_start in xrange(3):
tmpseq = fragseq[rf_start:].seq
if len(tmpseq) % 3:
tmpseq = tmpseq[:-(len(tmpseq) % 3)]
tmpprot = tmpseq.translate()
(score, ali1, ali2) = align_overlap(tmpprot, prot)
ali_score.append(score)
rf_start = np.argmax(ali_score)
positions = np.arange(frag_start + rf_start, frag_end) - frag_start
if len(positions) % 3:
positions = positions[:-(len(positions) % 3)]
return positions
def get_gene_positions_in_fragment(gene, fragment, gwseq, fragseq, VERBOSE=0):
"""Get the coordinates of a gene within a fragment"""
# Find coordinates of gene in reference
feagene = gwseq.features[map(attrgetter("id"), gwseq.features).index(gene)]
gene_start = feagene.location.nofuzzy_start
gene_end = feagene.location.nofuzzy_end
# Sanity check on coordinates
feafrag = refseq.features[map(attrgetter("id"), gwseq.features).index(fragment)]
fragrefgw = feafrag.extract(gwseq)
if len(fragseq) != len(fragrefgw):
raise ValueError("Problem with coordinates between fragment and genomewide.")
# Find coordinates of gene in fragment
frag_start = feafrag.location.nofuzzy_start
frag_end = feafrag.location.nofuzzy_end
# complete gene
if (frag_start <= gene_start) and (frag_end >= gene_end):
if VERBOSE >= 2:
print "Complete gene found"
positions = np.arange(gene_start, gene_end) - frag_start
# start of gene
elif frag_start <= gene_start:
if VERBOSE >= 2:
print "WARNING: only gene start found"
positions = np.arange(gene_start, frag_end) - frag_start
if len(positions) % 3:
positions = positions[: -(len(positions) % 3)]
# end of gene
elif frag_end >= gene_end:
if VERBOSE >= 2:
print "WARNING: only gene end found"
positions = np.arange(frag_start, gene_end) - frag_start
if len(positions) % 3:
positions = positions[len(positions) % 3 :]
# middle of gene: guess reading frame
else:
if VERBOSE >= 2:
print "WARNING: only gene middle found"
prot = feagene.extract(gwseq).seq.translate()
ali_score = []
for rf_start in xrange(3):
tmpseq = fragseq[rf_start:].seq
if len(tmpseq) % 3:
tmpseq = tmpseq[: -(len(tmpseq) % 3)]
tmpprot = tmpseq.translate()
(score, ali1, ali2) = align_overlap(tmpprot, prot)
ali_score.append(score)
rf_start = np.argmax(ali_score)
positions = np.arange(frag_start + rf_start, frag_end) - frag_start
if len(positions) % 3:
positions = positions[: -(len(positions) % 3)]
return positions
def trim_to_refseq(seq, refseq):
'''Trim sequence to a reference sequence'''
from seqanpy import align_overlap
(score, ali1, ali2) = align_overlap(seq, refseq, score_gapopen=-20)
start = len(ali2) - len(ali2.lstrip('-'))
end = len(ali2.rstrip('-'))
return seq[start: end]
def align_pairwise(seq1, seq2):
try:
from seqanpy import align_overlap
return align_overlap(seq1, seq2, **scoring_params)
except ImportError:
from Bio import pairwise2
aln = pairwise2.align.globalms(seq1, seq2,
scoring_params['score_match'], scoring_params['score_mismatch'],
scoring_params['score_gapopen'], scoring_params['score_gapext'],
penalize_end_gaps=False, one_alignment_only=True)[0]
return aln[2], aln[0], aln[1]
def align_dna(seqstr, refstr, require_full_cover=True):
if require_full_cover:
(score, alis, alir) = align_overlap(seqstr, refstr)
start = len(alir) - len(alir.lstrip('-'))
end = len(alir.rstrip('-'))
alist = alis[start: end]
alirt = alir[start: end]
else:
(score, alis, alir) = align_local(seqstr, refstr)
reftrim = alir.replace('-', '')
start = refstr.find(reftrim[:50])
end = refstr.rfind(reftrim[-50:]) + len(reftrim[-50:])
alist = ('N' * start) + alis + ('N' * (len(refstr) - end))
alirt = refstr[:start] + alir + refstr[end:]
return (alist, alirt)
def align_dna(seqstr, refstr, require_full_cover=True):
if require_full_cover:
(score, alis, alir) = align_overlap(seqstr, refstr)
start = len(alir) - len(alir.lstrip('-'))
end = len(alir.rstrip('-'))
alist = alis[start:end]
alirt = alir[start:end]
else:
(score, alis, alir) = align_local(seqstr, refstr)
reftrim = alir.replace('-', '')
start = refstr.find(reftrim[:50])
end = refstr.rfind(reftrim[-50:]) + len(reftrim[-50:])
alist = ('N' * start) + alis + ('N' * (len(refstr) - end))
alirt = refstr[:start] + alir + refstr[end:]
return (alist, alirt)
def get_distance_reads_sequence(seq, reads, VERBOSE=0,
score_match=3,
score_mismatch=-3):
'''Get the distance in alignment score between read pairs and a sequence'''
from seqanpy import align_overlap
seqs = ''.join(seq)
deltas = []
for irp, read_pair in enumerate(reads):
d = 0
for read in read_pair:
(score, alis, alir) = align_overlap(seqs, read.seq,
score_match=score_match,
score_mismatch=score_mismatch)
start = len(alir) - len(alir.lstrip('-'))
end = len(alir.rstrip('-'))
scoremax = score_match * (end - start)
delta = scoremax - score
d += delta
deltas.append(d)
return deltas
# Script
if __name__ == '__main__':
# Try import
import seqanpy as sap
# Global pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'AAATCGA'
output = sap.align_global(seq1, seq2, band=5)
print output
# Overlap pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'ATCT'
output = sap.align_overlap(seq1, seq2)
print output
# Overlap pairwise alignment cutting flanks
seq1 = 'AAAGGTCTA'
seq2 = 'ATCT'
output = sap.align_overlap(seq1, seq2, cut_flanks=True)
print output
# Ladder pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'TCTAGGGAAACCC'
output = sap.align_ladder(seq1, seq2)
print output
# Local pairwise alignment
def overlap_test():
print('Test align_overlap')
import seqanpy
(score, ali1, ali2) = seqanpy.align_overlap('ACCGT', 'CCG')
assert ali1 == 'ACCGT'
assert ali2 == '-CCG-'
scoremax = 3 * len(ali1)
delta = scoremax - score
ali = [ali2, ali1]
print 'Alignment to its own consensus (delta = '+str(delta)+')'
pretty_print_pairwise_ali(ali,
'cons',
'read'+str(i+1)+' '+str(edge),
len_name=25, width=90)
print ''
# Compare to all consensi and find the closest
alifr = alis[fragment]
alifrpw = []
for cons in alifr:
alifrpw.append(align_overlap(cons.seq.ungap('-'), seq))
scores = map(itemgetter(0), alifrpw)
indmax = np.argmax(scores)
alimax = alifrpw[indmax][1:]
start = len(alimax[1]) - len(alimax[1].lstrip('-'))
end = len(alimax[1].rstrip('-'))
alimax = [s[start: end] for s in alimax]
score = scores[indmax]
scoremax = 3 * len(alimax[0])
delta = scoremax - score
name1 = ' '.join(['cons '] + alifr[indmax].name.split('_')[::2])
name2 = ' '.join(['read'+str(i+1), pname, sample['patient sample']])
print 'Alignment to best consensus (delta = '+str(delta)+')'
pretty_print_pairwise_ali(alimax, name1, name2, len_name=25, width=90)
def get_minimal_distance_hist(bamfilename, consensi, maxreads=1000, VERBOSE=0):
'''Get histogram of minimal distance of reads from consensi'''
conssi = map(''.join, consensi)
m = np.zeros(len(consensi), int)
n_good = 0
with pysam.Samfile(bamfilename, 'rb') as bamfile:
for irp, reads in enumerate(pair_generator(bamfile)):
if n_good == maxreads:
break
if VERBOSE >= 3:
print n_good + 1, 'Checking mindist for:', reads[0].qname,
# Assign names
(read1, read2) = reads
i_fwd = reads[0].is_reverse
# Check a few things to make sure we are looking at paired reads
if read1.qname != read2.qname:
raise ValueError('Read pair '+str(irp)+': reads have different names!')
# Ignore unmapped reads
if read1.is_unmapped or read2.is_unmapped:
if VERBOSE >= 2:
print 'Read pair '+read1.qname+': unmapped'
continue
# Ignore not properly paired reads (this includes mates sitting on
# different fragments)
if (not read1.is_proper_pair) or (not read2.is_proper_pair):
if VERBOSE >= 2:
print 'Read pair '+read1.qname+': not properly paired'
continue
n_good += 1
# Get all distances
ds_pair = np.zeros_like(m)
for ic, consensus in enumerate(consensi):
conss = conssi[ic]
dpair = 0
for read in reads:
seq = read.seq
ali = align_overlap(conss, seq)
# NOTE: it is possible that we start before conss' start or end after
# its end, but that IS evidence that it's not contamination from there.
pos = conss.find(ali[1].replace('-', ''))
alim0 = np.fromstring(ali[1], 'S1')
alim1 = np.fromstring(ali[2], 'S1')
# Score subst
d = ((alim0 != alim1) & (alim0 != '-') & (alim1 != '-')).sum()
# Score insertions
gaps = alim0 == '-'
if gaps.sum():
n_gaps_borders = np.diff(gaps).sum()
n_gaps_borders += alim0[0] == '-'
n_gaps_borders += alim0[-1] == '-'
n_insertions = n_gaps_borders // 2
d += n_insertions
# Score deletions
gaps = alim1 == '-'
if gaps.sum():
n_gaps_borders = np.diff(gaps).sum()
n_gaps_borders -= alim1[0] == '-'
n_gaps_borders -= alim1[-1] == '-'
n_deletions = n_gaps_borders // 2
d += n_deletions
dpair += d
ds_pair[ic] = dpair
if VERBOSE >= 3:
print 'OK',
m[ds_pair.argmin()] += 1
if VERBOSE >= 3:
print ''
return m
def filter_contamination(
bamfilename,
bamfilename_out,
contseqs,
samplename,
VERBOSE=0,
deltascore_max_self=60,
deltascore_max_other=24,
maxreads=-1,
**kwargs
):
"""Fish contaminated reads from mapped reads
The function checks for a maximal distance to the expected consensus, and only
if it's more than that it checks all other samples.
Args:
deltascore_max_self (int): the maximal delta in alignment score to the
consensus to be considered pure
deltascore_max_other (int): the maximal delta in alignment score to any other
sample to be considered a contamination
**kwargs: passed down to the pairwise alignment function
"""
import pysam
from collections import defaultdict
from operator import itemgetter
from seqanpy import align_overlap
from hivwholeseq.utils.mapping import pair_generator, get_number_reads
if "score_match" in kwargs:
score_match = kwargs["score_match"]
else:
score_match = 3
bamfilename_trash = bamfilename_out[:-4] + "_trashed.bam"
contseqs = contseqs.copy()
consseq = contseqs.pop(samplename)
if VERBOSE >= 2:
print "Scanning reads (" + str(get_number_reads(bamfilename) // 2) + ")"
with pysam.Samfile(bamfilename, "rb") as bamfile:
with pysam.Samfile(bamfilename_out, "wb", template=bamfile) as bamfileout, pysam.Samfile(
bamfilename_trash, "wb", template=bamfile
) as bamfiletrash:
n_good = 0
n_cont = defaultdict(int)
for irp, reads in enumerate(pair_generator(bamfile)):
if irp == maxreads:
break
if VERBOSE >= 2:
if not ((irp + 1) % 100):
if not ((irp + 1) == 100):
sys.stdout.write("\x1b[1A")
print irp + 1
for read in reads:
# Look for distance to the own consensus, it that's small move on
alignments_read = {}
deltas_read = {}
(score, alis1, alis2) = align_overlap(consseq, read.seq, **kwargs)
(alis1, alis2) = trim_align_overlap((alis1, alis2))
scoremax = len(alis1) * score_match
delta_read = scoremax - score
deltas_read[samplename] = delta_read
alignments_read[samplename] = (alis1, alis2)
if delta_read <= deltascore_max_self:
if VERBOSE >= 4:
print "Read is very close to its own consensus", scoremax, score, delta_read
pretty_print_pairwise_ali([alis1, alis2], width=90, name1="ref", name2="read")
continue
# Otherwise, move on to all other sequences and find the neighbour
for contname, contseq in contseqs.iteritems():
(score, ali1, ali2) = align_overlap(contseq, read.seq, **kwargs)
(ali1, ali2) = trim_align_overlap((ali1, ali2))
scoremax = len(ali1) * score_match
delta_read = scoremax - score
deltas_read[contname] = delta_read
alignments_read[contname] = (ali1, ali2)
if VERBOSE >= 5:
print samplename
for key, d in deltas_read.iteritems():
print key, d
(contname, delta_read) = min(deltas_read.iteritems(), key=itemgetter(1))
# Again, the correct consensus has precedence
if deltas_read[samplename] == delta_read:
contname = samplename
(ali1, ali2) = alignments_read[contname]
# The read may be closest to its own consensus, if not very close
if contname == samplename:
if VERBOSE >= 4:
print "Read is closest to its consensus", scoremax, score, delta_read
pretty_print_pairwise_ali([ali1, ali2], width=90, name1="ref", name2="read")
# The read may come from another consensus (contamination)
elif delta_read <= deltascore_max_other:
n_cont[contname] += 1
bamfiletrash.write(reads[0])
bamfiletrash.write(reads[1])
if VERBOSE >= 2:
print "Contaminated read found! Good:", n_good, "cont:", sum(
n_cont.itervalues()
), "sources:", n_cont
if VERBOSE >= 3:
print "Read is contaminated by", contname, scoremax, score, delta_read
pretty_print_pairwise_ali([alis1, alis2], width=90, name1="self", name2="read")
print ""
pretty_print_pairwise_ali([ali1, ali2], width=90, name1="ref", name2="read")
if VERBOSE >= 2:
print ""
break
# Finally, the read is not really close to anything: accept
else:
if VERBOSE >= 4:
print "Read is close to nothing really", scoremax, score, delta_read
pretty_print_pairwise_ali([ali1, ali2], width=90, name1="ref", name2="read")
else:
n_good += 1
bamfileout.write(reads[0])
bamfileout.write(reads[1])
n_cont = dict(n_cont)
return (n_good, n_cont)
def merge_allele_counts(ref_genomewide, acs, VERBOSE=0):
'''Merge the allele counts of all fragments
Note: we do not require full coverage of all fragments, the missing
ones will just have zero counts. Sometimes, cherry-picking the data
fragment by fragment might be a better choice.
'''
from hivwholeseq.utils.miseq import alpha, read_types
from seqanpy import align_overlap
ac = np.zeros((len(read_types), len(alpha), len(ref_genomewide)), int)
pos_ref = 1000
for (fr, ref, acsi) in acs:
# Find the coordinates
(score, ali1, ali2) = align_overlap(
ref_genomewide[pos_ref - 1000:],
ref,
#score_gapopen=-20,
)
fr_start = len(ali2) - len(ali2.lstrip('-'))
fr_end = len(ali2.rstrip('-'))
if VERBOSE:
print fr, pos_ref - 1000 + fr_start, pos_ref - 1000 + fr_end
# Scan the alignment
pos_ref = pos_ref - 1000 + fr_start
fr_start_ref = pos_ref
fr_end_ref = pos_ref + fr_end - fr_start
pos_fr = 0
for pos_ali in xrange(fr_start, fr_end):
# Gap in genomewise, ignore position
if ali1[pos_ali] == '-':
pos_fr += 1
continue
# Gap in fragment, ignore FIXME: probably we should put deletions
elif ali2[pos_ali] == '-':
pos_ref += 1
continue
# Add the counts
# NOTE: all fragments are treated the same, even in case of coverage
# differences of orders of magnitude. This means, larger coverage
# always wins. Maybe we want to implement this somewhat differently
ac[:, :, pos_ref] += acsi[:, :, pos_fr]
pos_fr += 1
pos_ref += 1
if VERBOSE >= 3:
from hivwholeseq.utils.sequence import pretty_print_pairwise_ali
cons = alpha[ac.sum(axis=0).argmax(axis=0)]
pretty_print_pairwise_ali(
(ali1[fr_start:fr_end], cons[fr_start:fr_end]),
name1='gw',
name2=fr,
width=100)
return ac
def annotate_sequence(seqrecord,
additional_edges={},
additional_features=['chunk'],
VERBOSE=0):
'''Annotate a consensus with the genes and stuff (in place)'''
# TODO: what do we do with genes that do not start/end where they are
# supposed to? Do we follow biology and track their new locations?
from Bio.SeqFeature import SeqFeature, FeatureLocation, CompoundLocation
from hivwholeseq.utils.genome_info import gene_edges, RNA_structure_edges, \
other_edges, find_region_edges, find_region_edges_multiple, \
locate_gene
edge_dict = {
'gene': gene_edges,
'RNA structure': RNA_structure_edges,
'other': other_edges
}
edge_dict.update(additional_edges)
additional_features = ['protein'] + additional_features
features = edge_dict.keys() + additional_features
if VERBOSE:
print 'Features:', ', '.join(features)
smat = np.array(seqrecord)
for feature_type in edge_dict:
edges_all = edge_dict[feature_type]
print feature_type, edge_dict[feature_type].keys()
for name, edges in edges_all.iteritems():
if VERBOSE >= 2:
print name,
# Skip a feature if it's present already
if name in map(lambda x: x.id, seqrecord.features):
if VERBOSE >= 2:
print 'already present.'
continue
# Behave differently for unsplit regions and split ones
if len(edges) == 2:
# LTR problems with F6
if 'F6' in name:
pos_edge = find_region_edges(smat[6000::],
[edges[0], None])
pos_edge[0] += 6000
elif feature_type == 'genes':
pos_edge = locate_gene(smat, name, output_compact=True)
else:
pos_edge = find_region_edges(smat, edges)
# Cut the primers for some features
if (None not in pos_edge) and name in ['V1', 'V3', 'V4', 'V5']:
pos_edge[0] += len(edges[0])
pos_edge[1] -= len(edges[1])
# Cut only the right primer for V2
if (None not in pos_edge) and name in ['V2']:
pos_edge[1] -= len(edges[1])
if pos_edge[0] is None:
if name not in ['F1', "LTR5'"]:
print 'WARNING: start not found'
pos_edge[0] = 0
if pos_edge[1] is None:
if name not in ['F6', "LTR3'"]:
print 'WARNING: end not found'
pos_edge[1] = len(smat)
location = FeatureLocation(*pos_edge)
else:
if feature_type == 'genes':
pos_edges = [
locate_gene(smat, name + suff, output_compact=True)
for suff in ('1', '2')
]
else:
pos_edges = find_region_edges_multiple(smat,
edges,
min_distance=1)
locations = [
FeatureLocation(*pos_edge) for pos_edge in pos_edges
]
location = CompoundLocation(locations)
if VERBOSE >= 2:
print 'found:', location
feature = SeqFeature(location,
type=feature_type,
id=name,
strand=1)
seqrecord.features.append(feature)
# Add proteins and other features from HXB2
from operator import attrgetter
from seqanpy import align_overlap
from hivwholeseq.utils.genome_info import proteins, chunks
from hivwholeseq.reference import load_custom_reference
additional_features_dict = {}
if 'protein' in additional_features:
additional_features_dict['protein'] = proteins
if 'chunk' in additional_features:
additional_features_dict['chunk'] = chunks
ref_ann = load_custom_reference('HXB2', 'gb')
for feagroup, additional_features_grp in additional_features_dict.iteritems(
):
for feaname in additional_features_grp:
if VERBOSE >= 2:
print feaname,
fea = ref_ann.features[map(attrgetter('id'),
ref_ann.features).index(feaname)]
seq = fea.extract(ref_ann)
(score, ali1, ali2) = align_overlap(seqrecord,
seq,
score_gapopen=-20)
start = len(ali2) - len(ali2.lstrip('-'))
end = len(ali2.rstrip('-'))
end -= ali1[start:end].count('-')
location = FeatureLocation(start, end)
if VERBOSE >= 2:
print 'found:', location
feature = SeqFeature(location, type=feagroup, id=feaname, strand=1)
seqrecord.features.append(feature)
def filter_contamination(bamfilename,
bamfilename_out,
contseqs,
samplename,
VERBOSE=0,
deltascore_max_self=60,
deltascore_max_other=24,
maxreads=-1,
**kwargs):
'''Fish contaminated reads from mapped reads
The function checks for a maximal distance to the expected consensus, and only
if it's more than that it checks all other samples.
Args:
deltascore_max_self (int): the maximal delta in alignment score to the
consensus to be considered pure
deltascore_max_other (int): the maximal delta in alignment score to any other
sample to be considered a contamination
**kwargs: passed down to the pairwise alignment function
'''
import pysam
from collections import defaultdict
from operator import itemgetter
from seqanpy import align_overlap
from hivwholeseq.utils.mapping import pair_generator, get_number_reads
if 'score_match' in kwargs:
score_match = kwargs['score_match']
else:
score_match = 3
bamfilename_trash = bamfilename_out[:-4] + '_trashed.bam'
contseqs = contseqs.copy()
consseq = contseqs.pop(samplename)
if VERBOSE >= 2:
print 'Scanning reads (' + str(
get_number_reads(bamfilename) // 2) + ')'
with pysam.Samfile(bamfilename, 'rb') as bamfile:
with pysam.Samfile(bamfilename_out, 'wb', template=bamfile) as bamfileout, \
pysam.Samfile(bamfilename_trash, 'wb', template=bamfile) as bamfiletrash:
n_good = 0
n_cont = defaultdict(int)
for irp, reads in enumerate(pair_generator(bamfile)):
if irp == maxreads:
break
if VERBOSE >= 2:
if not ((irp + 1) % 100):
if not ((irp + 1) == 100):
sys.stdout.write('\x1b[1A')
print irp + 1
for read in reads:
# Look for distance to the own consensus, it that's small move on
alignments_read = {}
deltas_read = {}
(score, alis1,
alis2) = align_overlap(consseq, read.seq, **kwargs)
(alis1, alis2) = trim_align_overlap((alis1, alis2))
scoremax = len(alis1) * score_match
delta_read = scoremax - score
deltas_read[samplename] = delta_read
alignments_read[samplename] = (alis1, alis2)
if delta_read <= deltascore_max_self:
if VERBOSE >= 4:
print 'Read is very close to its own consensus', scoremax, score, delta_read
pretty_print_pairwise_ali([alis1, alis2],
width=90,
name1='ref',
name2='read')
continue
# Otherwise, move on to all other sequences and find the neighbour
for contname, contseq in contseqs.iteritems():
(score, ali1,
ali2) = align_overlap(contseq, read.seq, **kwargs)
(ali1, ali2) = trim_align_overlap((ali1, ali2))
scoremax = len(ali1) * score_match
delta_read = scoremax - score
deltas_read[contname] = delta_read
alignments_read[contname] = (ali1, ali2)
if VERBOSE >= 5:
print samplename
for key, d in deltas_read.iteritems():
print key, d
(contname, delta_read) = min(deltas_read.iteritems(),
key=itemgetter(1))
# Again, the correct consensus has precedence
if deltas_read[samplename] == delta_read:
contname = samplename
(ali1, ali2) = alignments_read[contname]
# The read may be closest to its own consensus, if not very close
if contname == samplename:
if VERBOSE >= 4:
print 'Read is closest to its consensus', scoremax, score, delta_read
pretty_print_pairwise_ali([ali1, ali2],
width=90,
name1='ref',
name2='read')
# The read may come from another consensus (contamination)
elif (delta_read <= deltascore_max_other):
n_cont[contname] += 1
bamfiletrash.write(reads[0])
bamfiletrash.write(reads[1])
if VERBOSE >= 2:
print 'Contaminated read found! Good:', n_good, 'cont:', sum(
n_cont.itervalues()), 'sources:', n_cont
if VERBOSE >= 3:
print 'Read is contaminated by', contname, scoremax, score, delta_read
pretty_print_pairwise_ali([alis1, alis2],
width=90,
name1='self',
name2='read')
print ''
pretty_print_pairwise_ali([ali1, ali2],
width=90,
name1='ref',
name2='read')
if VERBOSE >= 2:
print ''
break
# Finally, the read is not really close to anything: accept
else:
if VERBOSE >= 4:
print 'Read is close to nothing really', scoremax, score, delta_read
pretty_print_pairwise_ali([ali1, ali2],
width=90,
name1='ref',
name2='read')
else:
n_good += 1
bamfileout.write(reads[0])
bamfileout.write(reads[1])
n_cont = dict(n_cont)
return (n_good, n_cont)
# Script
if __name__ == '__main__':
# Try import
import seqanpy as sap
# Global pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'AAATCGA'
output = sap.align_global(seq1, seq2, band=5)
print output
# Overlap pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'ATCT'
output = sap.align_overlap(seq1, seq2)
print output
# Ladder pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'TCTAGGGAAACCC'
output = sap.align_ladder(seq1, seq2)
print output
# Local pairwise alignment
seq1 = 'AAAGGTCTACCGTAGCCT'
seq2 = 'AAGTCTAC'
output = sap.align_local(seq1, seq2)
print output
def annotate_sequence(seqrecord, additional_edges={}, additional_features=['chunk'], VERBOSE=0):
'''Annotate a consensus with the genes and stuff (in place)'''
# TODO: what do we do with genes that do not start/end where they are
# supposed to? Do we follow biology and track their new locations?
from Bio.SeqFeature import SeqFeature, FeatureLocation, CompoundLocation
from hivwholeseq.utils.genome_info import gene_edges, RNA_structure_edges, \
other_edges, find_region_edges, find_region_edges_multiple, \
locate_gene
edge_dict = {'gene': gene_edges,
'RNA structure': RNA_structure_edges,
'other': other_edges}
edge_dict.update(additional_edges)
additional_features = ['protein'] + additional_features
features = edge_dict.keys() + additional_features
if VERBOSE:
print 'Features:', ', '.join(features)
smat = np.array(seqrecord)
for feature_type in edge_dict:
edges_all = edge_dict[feature_type]
print feature_type, edge_dict[feature_type].keys()
for name, edges in edges_all.iteritems():
if VERBOSE >= 2:
print name,
# Skip a feature if it's present already
if name in map(lambda x: x.id, seqrecord.features):
if VERBOSE >= 2:
print 'already present.'
continue
# Behave differently for unsplit regions and split ones
if len(edges) == 2:
# LTR problems with F6
if 'F6' in name:
pos_edge = find_region_edges(smat[6000::], [edges[0], None])
pos_edge[0] += 6000
elif feature_type == 'genes':
pos_edge = locate_gene(smat, name, output_compact=True)
else:
pos_edge = find_region_edges(smat, edges)
# Cut the primers for some features
if (None not in pos_edge) and name in ['V1', 'V3', 'V4', 'V5']:
pos_edge[0] += len(edges[0])
pos_edge[1] -= len(edges[1])
# Cut only the right primer for V2
if (None not in pos_edge) and name in ['V2']:
pos_edge[1] -= len(edges[1])
if pos_edge[0] is None:
if name not in ['F1', "LTR5'"]:
print 'WARNING: start not found'
pos_edge[0] = 0
if pos_edge[1] is None:
if name not in ['F6', "LTR3'"]:
print 'WARNING: end not found'
pos_edge[1] = len(smat)
location = FeatureLocation(*pos_edge)
else:
if feature_type == 'genes':
pos_edges = [locate_gene(smat, name+suff, output_compact=True)
for suff in ('1', '2')]
else:
pos_edges = find_region_edges_multiple(smat, edges, min_distance=1)
locations = [FeatureLocation(*pos_edge) for pos_edge in pos_edges]
location = CompoundLocation(locations)
if VERBOSE >= 2:
print 'found:', location
feature = SeqFeature(location, type=feature_type, id=name, strand=1)
seqrecord.features.append(feature)
# Add proteins and other features from HXB2
from operator import attrgetter
from seqanpy import align_overlap
from hivwholeseq.utils.genome_info import proteins, chunks
from hivwholeseq.reference import load_custom_reference
additional_features_dict = {}
if 'protein' in additional_features:
additional_features_dict['protein'] = proteins
if 'chunk' in additional_features:
additional_features_dict['chunk'] = chunks
ref_ann = load_custom_reference('HXB2', 'gb')
for feagroup, additional_features_grp in additional_features_dict.iteritems():
for feaname in additional_features_grp:
if VERBOSE >= 2:
print feaname,
fea = ref_ann.features[map(attrgetter('id'), ref_ann.features).index(feaname)]
seq = fea.extract(ref_ann)
(score, ali1, ali2) = align_overlap(seqrecord, seq, score_gapopen=-20)
start = len(ali2) - len(ali2.lstrip('-'))
end = len(ali2.rstrip('-'))
end -= ali1[start: end].count('-')
location = FeatureLocation(start, end)
if VERBOSE >= 2:
print 'found:', location
feature = SeqFeature(location, type=feagroup, id=feaname, strand=1)
seqrecord.features.append(feature)
]
seq = ''.join(seql)
s, a1, a2 = align_local(prot['seq_aa'], seq)
scores.append(s)
chain = list(struc.get_chains())[np.argmax(s)]
seql = [d3to1.get(r.get_resname(), 'O') for r in chain.get_residues()]
seq = ''.join(seql)
# Flag all mutations
for m, mut in muts.iterrows():
mutations.at[m, 'PDB_fn'] = fns[protname]
mutations.at[m, 'PDB_id'] = fns[protname].split(
'_')[1].upper().split('.')[0]
mutations.at[m, 'PDB_chain'] = chain.id
s, a1, a2 = align_overlap(seq, mut['context_protein'])
# The focal allele is always small and is the only such letter
pos_in_context = 4
pos = a2.find(mut['context_protein'][pos_in_context])
pos -= a1[:pos].count('-')
mutations.at[m, 'PDB_pos_in_chain'] = pos
mutations.at[m, 'PDB_allele'] = seq[pos]
print('Ref: ' + mut['context_protein'])
print('PDB: ' + seq[pos - 4:pos] + seq[pos].lower() +
seq[pos + 1:pos + 5])
#mutations.to_csv('../data/mutations_highvariance_summary.tsv', sep='\t', index=True)
print('Load multiple sequence alignments')
from Bio import AlignIO
protname = 'NS1'
def get_minimal_distance_hist(bamfilename, consensi, maxreads=1000, VERBOSE=0):
'''Get histogram of minimal distance of reads from consensi'''
conssi = map(''.join, consensi)
m = np.zeros(len(consensi), int)
n_good = 0
with pysam.Samfile(bamfilename, 'rb') as bamfile:
for irp, reads in enumerate(pair_generator(bamfile)):
if n_good == maxreads:
break
if VERBOSE >= 3:
print n_good + 1, 'Checking mindist for:', reads[0].qname,
# Assign names
(read1, read2) = reads
i_fwd = reads[0].is_reverse
# Check a few things to make sure we are looking at paired reads
if read1.qname != read2.qname:
raise ValueError('Read pair ' + str(irp) +
': reads have different names!')
# Ignore unmapped reads
if read1.is_unmapped or read2.is_unmapped:
if VERBOSE >= 2:
print 'Read pair ' + read1.qname + ': unmapped'
continue
# Ignore not properly paired reads (this includes mates sitting on
# different fragments)
if (not read1.is_proper_pair) or (not read2.is_proper_pair):
if VERBOSE >= 2:
print 'Read pair ' + read1.qname + ': not properly paired'
continue
n_good += 1
# Get all distances
ds_pair = np.zeros_like(m)
for ic, consensus in enumerate(consensi):
conss = conssi[ic]
dpair = 0
for read in reads:
seq = read.seq
ali = align_overlap(conss, seq)
# NOTE: it is possible that we start before conss' start or end after
# its end, but that IS evidence that it's not contamination from there.
pos = conss.find(ali[1].replace('-', ''))
alim0 = np.fromstring(ali[1], 'S1')
alim1 = np.fromstring(ali[2], 'S1')
# Score subst
d = ((alim0 != alim1) & (alim0 != '-') &
(alim1 != '-')).sum()
# Score insertions
gaps = alim0 == '-'
if gaps.sum():
n_gaps_borders = np.diff(gaps).sum()
n_gaps_borders += alim0[0] == '-'
n_gaps_borders += alim0[-1] == '-'
n_insertions = n_gaps_borders // 2
d += n_insertions
# Score deletions
gaps = alim1 == '-'
if gaps.sum():
n_gaps_borders = np.diff(gaps).sum()
n_gaps_borders -= alim1[0] == '-'
n_gaps_borders -= alim1[-1] == '-'
n_deletions = n_gaps_borders // 2
d += n_deletions
dpair += d
ds_pair[ic] = dpair
if VERBOSE >= 3:
print 'OK',
m[ds_pair.argmin()] += 1
if VERBOSE >= 3:
print ''
return m
# Script
if __name__ == '__main__':
# Try import
import seqanpy as sap
# Global pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'AAATCGA'
output = sap.align_global(seq1, seq2, band=5)
print output
# Overlap pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'ATCT'
output = sap.align_overlap(seq1, seq2)
print output
# Ladder pairwise alignment
seq1 = 'AAAGGTCTA'
seq2 = 'TCTAGGGAAACCC'
output = sap.align_ladder(seq1, seq2)
print output
# Local pairwise alignment
seq1 = 'AAAGGTCTACCGTAGCCT'
seq2 = 'AAGTCTAC'
output = sap.align_local(seq1, seq2)
print output
patient = load_patient(pname)
refseqgw = patient.get_reference('genomewide')
for fragment in fragments:
if VERBOSE >= 1:
print pname, fragment
if VERBOSE >= 2:
print 'Cutting out fragment', fragment
# Get start coordinate
if fragment == 'F1':
start = 0
else:
prfwd = primers_outer[fragment][0]
(score, ali1, ali2) = align_overlap(refseqgw, prfwd, score_gapopen=-20)
start = len(ali2) - len(ali2.lstrip('-')) + len(prfwd)
# Get end coordinate
if fragment == 'F6':
end = len(refseqgw)
else:
prrev = primers_outer[fragment][1]
(score, ali1, ali2) = align_overlap(refseqgw, prrev, score_gapopen=-20)
end = len(ali2) - len(ali2.lstrip('-'))
refseq = refseqgw[start: end]
refseq.id = patient.code+'_ref_'+fragment
refseq.name = refseq.id
refseq.description = 'Patient '+patient.code+', initial reference '+fragment