def get_cds(self, seq_dict):
"""
Return the CDS sequence (as a string) for the transcript
(based on the exons) using a sequenceDict as the sequence source.
The returned sequence is in the correct 5'-3' orientation (i.e. it has
been reverse complemented if necessary).
"""
sequence = seq_dict[self.chromosome]
assert self.stop <= len(sequence) + 1
# make sure this isn't a non-coding gene
if self.thick_start == self.thick_stop == 0:
return ''
s = []
for e in self.exon_intervals:
if self.thick_start < e.start and e.stop < self.thick_stop:
# squarely in the CDS
s.append(sequence[e.start:e.stop])
elif e.start <= self.thick_start < e.stop < self.thick_stop:
# thickStart marks the start of the CDS
s.append(sequence[self.thick_start:e.stop])
elif e.start <= self.thick_start and self.thick_stop <= e.stop:
# thickStart and thickStop mark the whole CDS
s.append(sequence[self.thick_start: self.thick_stop])
elif self.thick_start < e.start < self.thick_stop <= e.stop:
# thickStop marks the end of the CDS
s.append(sequence[e.start:self.thick_stop])
if self.strand == '-':
cds = reverse_complement(''.join(s))
else:
cds = ''.join(s)
return str(cds)
def get_cds(self, seq_dict):
"""
Return the CDS sequence (as a string) for the transcript
(based on the exons) using a sequenceDict as the sequence source.
The returned sequence is in the correct 5'-3' orientation (i.e. it has
been reverse complemented if necessary).
"""
sequence = seq_dict[self.chromosome]
assert self.stop <= len(sequence) + 1
# make sure this isn't a non-coding gene
if self.thick_start == self.thick_stop == 0:
return ''
s = []
for e in self.exon_intervals:
if self.thick_start < e.start and e.stop < self.thick_stop:
# squarely in the CDS
s.append(sequence[e.start:e.stop])
elif e.start <= self.thick_start < e.stop < self.thick_stop:
# thickStart marks the start of the CDS
s.append(sequence[self.thick_start:e.stop])
elif e.start <= self.thick_start and self.thick_stop <= e.stop:
# thickStart and thickStop mark the whole CDS
s.append(sequence[self.thick_start:self.thick_stop])
elif self.thick_start < e.start < self.thick_stop <= e.stop:
# thickStop marks the end of the CDS
s.append(sequence[e.start:self.thick_stop])
if self.strand == '-':
cds = reverse_complement(''.join(s))
else:
cds = ''.join(s)
return str(cds)
def get_sequence(self, seq_dict, stranded=True):
"""
Returns the sequence for this ChromosomeInterval. If stranded is True, reverse complements as necessary.
:param seq_dict: Dictionary-like object with DNA sequences.
:param stranded: Should we reverse complement negative strand sequences?
:return: A sequence string.
"""
if stranded is False or self.strand is '+':
return seq_dict[self.chromosome][self.start: self.stop]
elif self.strand is '-':
return reverse_complement(seq_dict[self.chromosome][self.start: self.stop])
def get_sequence(self, seq_dict, stranded=True):
"""
Returns the sequence for this ChromosomeInterval. If stranded is True, reverse complements as necessary.
:param seq_dict: Dictionary-like object with DNA sequences.
:param stranded: Should we reverse complement negative strand sequences?
:return: A sequence string.
"""
if stranded is False or self.strand is '+':
return seq_dict[self.chromosome][self.start:self.stop]
elif self.strand is '-':
return reverse_complement(
seq_dict[self.chromosome][self.start:self.stop])
def get_mrna(self, seq_dict):
"""
Returns the mRNA sequence for this transcript based on a Fasta object.
and the start/end positions and the exons. Sequence returned in
5'-3' transcript orientation.
"""
sequence = seq_dict[self.chromosome]
assert self.stop <= len(sequence) + 1
s = []
for e in self.exon_intervals:
s.append(sequence[e.start:e.stop])
if self.strand == '+':
mrna = ''.join(s)
else:
mrna = reverse_complement(''.join(s))
return str(mrna)
def get_mrna(self, seq_dict):
"""
Returns the mRNA sequence for this transcript based on a Fasta object.
and the start/end positions and the exons. Sequence returned in
5'-3' transcript orientation.
"""
sequence = seq_dict[self.chromosome]
assert self.stop <= len(sequence) + 1
s = []
for e in self.exon_intervals:
s.append(sequence[e.start:e.stop])
if self.strand == '+':
mrna = ''.join(s)
else:
mrna = reverse_complement(''.join(s))
return str(mrna)
def _write( self, pos, sequence, quality, variation_map, target_fh, variations=set(), offset=0, length=0, debug='', inversion=False ):
'''
write a sequence of given quality
debug is not used and for humans only
'''
if inversion:
inversion_text = 'inversion_'
sequence = bio.reverse_complement( sequence )
quality = quality[::-1]
else:
inversion_text = ''
target_fh.write( '@mgsa_seq_%i~%i~%i\n' % ( pos, offset, length ) ) # sam is 0 indexed
if len(variations) > 0:
variation_map.write( '@mgsa_seq_%i~%i~%i: %s_%s%s\n' % ( pos, offset, length, ','.join([ v for v in variations ]), inversion_text, debug ) ) # sam is 0 indexed
target_fh.write( sequence )
target_fh.write( '\n+\n' )
target_fh.write( quality ) # quality
target_fh.write( '\n' )
import sys
from bio import reverse_complement
with open(sys.argv[1]) as f:
read_data = f.read().strip()
print(reverse_complement(read_data))
def main(args):
used_only, args = grace.get_option_value(args, '--used-only',
grace.as_bool, False)
grace.expect_no_further_options(args)
if len(args) != 1:
sys.stderr.write(USAGE)
return 1
working_dir = args[0]
print
print 'Note: This is still under development'
print ' Pairing information is not included'
print ' Only the part of the read that was aligned is included'
print
if used_only:
hit_filename = 'used_shrimp_hits.txt.gz'
output_prefix = 'used_hits'
else:
hit_filename = 'shrimp_hits.txt.gz'
output_prefix = 'hits'
reference_filename = os.path.join(working_dir, 'reference.fa')
references = dict(io.read_fasta(reference_filename))
for name in references:
references[name] = references[name].upper()
bam_filename = safe_filename(working_dir, output_prefix + '_unsorted.bam')
bam_sorted_prefix = safe_filename(working_dir, output_prefix)
f = open(bam_filename, 'wb')
sam_eater = run(['samtools', 'view', '-S', '-b', '-'],
stdin=subprocess.PIPE,
stdout=f.fileno())
f.close()
for name in references:
print >> sam_eater.stdin, '@SQ\tSN:%s\tLN:%d' % (name,
len(references[name]))
for i, (read, hits) in enumerate(
shrimp.iter_read_hits(working_dir, hit_filename, qualities=True)):
if (i % 10000) == 0:
grace.status('Processing read %s' % grace.pretty_number(i))
for line in hits:
parts = line.rstrip('\n').split('\t')
read_name = parts[0]
ref_name = parts[1]
ref_start = int(parts[3]) - 1
ref_end = int(parts[4])
read_start = int(parts[5]) - 1
read_end = int(parts[6])
read_length = int(parts[7])
score = int(parts[8])
forward = (parts[2] == '+')
edit_string = parts[9]
corresp_seq = references[ref_name][ref_start:ref_end]
if not forward:
corresp_seq = bio.reverse_complement(corresp_seq)
hit_ref_ali, hit_read_ali = consensus.edit_string_to_alignment(
edit_string, corresp_seq)
if not forward:
hit_ref_ali = bio.reverse_complement(hit_ref_ali)
hit_read_ali = bio.reverse_complement(hit_read_ali)
#Normalization -- move "-"s as far right as possible
hit_read_ali = consensus.roll_alignment(hit_read_ali, hit_ref_ali)
hit_ref_ali = consensus.roll_alignment(hit_ref_ali, hit_read_ali)
if len(read) > 2:
qual = read[2][read_start:read_end]
else:
qual = '*'
if forward:
forwardized_seq = read[1][read_start:read_end]
forwardized_qual = qual
else:
forwardized_seq = bio.reverse_complement(
read[1][read_start:read_end])
forwardized_qual = qual[::-1]
#if forwardized_seq != hit_read_ali.replace('-',''):
# print line
# print read
# print corresp_seq
assert forwardized_seq == hit_read_ali.replace(
'-', ''), forwardized_seq + ' ' + hit_read_ali
cigar_items = []
for i in xrange(len(hit_ref_ali)):
if hit_ref_ali[i] == '-':
cigar_items.append('I')
elif hit_read_ali[i] == '-':
cigar_items.append('D')
else:
cigar_items.append('M')
cigar = ''.join(
str(len(list(subiterator))) + key
for key, subiterator in itertools.groupby(cigar_items))
flags = 0
if not forward: flags += 0x0010
sam_line = '\t'.join([
read_name,
str(flags),
ref_name,
str(ref_start + 1),
'255',
cigar,
'=',
'0',
'0',
forwardized_seq,
forwardized_qual,
])
print >> sam_eater.stdin, sam_line
sam_eater.stdin.close()
assert sam_eater.wait() == 0, 'samtools failed'
grace.status('')
execute(['samtools', 'sort', bam_filename, bam_sorted_prefix])
os.unlink(bam_filename)
execute(['samtools', 'index', bam_sorted_prefix + '.bam'])
print
print working_dir + '/' + output_prefix + '.bam and index created'
print
def main(args):
used_only, args = grace.get_option_value(args,'--used-only', grace.as_bool, False)
grace.expect_no_further_options(args)
if len(args) != 1:
sys.stderr.write( USAGE )
return 1
working_dir = args[0]
print
print 'Note: This is still under development'
print ' Pairing information is not included'
print ' Only the part of the read that was aligned is included'
print
if used_only:
hit_filename = 'used_shrimp_hits.txt.gz'
output_prefix = 'used_hits'
else:
hit_filename = 'shrimp_hits.txt.gz'
output_prefix = 'hits'
reference_filename = os.path.join(working_dir,'reference.fa')
references = dict( io.read_fasta(reference_filename) )
for name in references:
references[name] = references[name].upper()
bam_filename = safe_filename(working_dir, output_prefix+'_unsorted.bam')
bam_sorted_prefix = safe_filename(working_dir, output_prefix)
f = open(bam_filename, 'wb')
sam_eater = run(['samtools', 'view', '-S', '-b', '-'],
stdin=subprocess.PIPE,
stdout=f.fileno())
f.close()
for name in references:
print >> sam_eater.stdin, '@SQ\tSN:%s\tLN:%d' % (name, len(references[name]))
for i, (read, hits) in enumerate(shrimp.iter_read_hits(working_dir, hit_filename, qualities=True)):
if (i % 10000) == 0:
grace.status('Processing read %s' % grace.pretty_number(i))
for line in hits:
parts = line.rstrip('\n').split('\t')
read_name = parts[0]
ref_name = parts[1]
ref_start = int(parts[3])-1
ref_end = int(parts[4])
read_start = int(parts[5])-1
read_end = int(parts[6])
read_length = int(parts[7])
score = int(parts[8])
forward = (parts[2] == '+')
edit_string = parts[9]
corresp_seq = references[ref_name][ref_start:ref_end]
if not forward:
corresp_seq = bio.reverse_complement(corresp_seq)
hit_ref_ali, hit_read_ali = consensus.edit_string_to_alignment(edit_string, corresp_seq)
if not forward:
hit_ref_ali = bio.reverse_complement(hit_ref_ali)
hit_read_ali = bio.reverse_complement(hit_read_ali)
#Normalization -- move "-"s as far right as possible
hit_read_ali = consensus.roll_alignment(hit_read_ali, hit_ref_ali)
hit_ref_ali = consensus.roll_alignment(hit_ref_ali, hit_read_ali)
if len(read) > 2:
qual = read[2][read_start:read_end]
else:
qual = '*'
if forward:
forwardized_seq = read[1][read_start:read_end]
forwardized_qual = qual
else:
forwardized_seq = bio.reverse_complement(read[1][read_start:read_end])
forwardized_qual = qual[::-1]
#if forwardized_seq != hit_read_ali.replace('-',''):
# print line
# print read
# print corresp_seq
assert forwardized_seq == hit_read_ali.replace('-',''), forwardized_seq + ' ' + hit_read_ali
cigar_items = [ ]
for i in xrange(len(hit_ref_ali)):
if hit_ref_ali[i] == '-':
cigar_items.append('I')
elif hit_read_ali[i] == '-':
cigar_items.append('D')
else:
cigar_items.append('M')
cigar = ''.join(
str(len(list(subiterator))) + key
for key, subiterator in itertools.groupby(cigar_items)
)
flags = 0
if not forward: flags += 0x0010
sam_line = '\t'.join([
read_name,
str(flags),
ref_name,
str(ref_start+1),
'255',
cigar,
'=',
'0',
'0',
forwardized_seq,
forwardized_qual,
])
print >> sam_eater.stdin, sam_line
sam_eater.stdin.close()
assert sam_eater.wait() == 0, 'samtools failed'
grace.status('')
execute([
'samtools', 'sort', bam_filename, bam_sorted_prefix
])
os.unlink(bam_filename)
execute([
'samtools', 'index', bam_sorted_prefix + '.bam'
])
print
print working_dir + '/' + output_prefix + '.bam and index created'
print