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)
# 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 not self.strand:
cds = reverse_complement("".join(s))
else:
cds = "".join(s)
return cds
def get_sequence(self, seq_dict, stranded=True):
"""
Returns the sequence for this intron in transcript orientation (reverse complement as necessary)
If strand is False, returns the + strand regardless of transcript orientation.
"""
if stranded is False or self.strand is True:
return seq_dict[self.chromosome][self.start:self.stop]
if self.strand is False:
return reverse_complement(seq_dict[self.chromosome][self.start:self.stop])
def bam_to_rec(read):
"""Convert pysam record to fastq
https://www.biostars.org/p/6970/
"""
seq = read.seq
qual = read.qual
if read.is_reverse:
seq = reverse_complement(seq)
qual = qual[::-1]
return "@{}\n{}\n+\n{}\n".format(read.qname, seq, qual)
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)
s = []
for e in self.exon_intervals:
s.append(sequence[e.start:e.stop])
if self.strand is True:
mrna = "".join(s)
else:
mrna = reverse_complement("".join(s))
return mrna
def get_cds(self, seq_dict, in_frame=True):
"""
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).
Overrides get_cds in GenePredTranscript to provide frame functionality
TODO: now loses the store-and-reuse functionality to avoid slicing offset sizes each time.
"""
sequence = seq_dict[self.chromosome]
assert self.stop <= len(sequence)
# 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 not self.strand:
cds = reverse_complement("".join(s))
else:
cds = "".join(s)
if in_frame is True:
offset = find_offset(self.exon_frames, self.strand)
cds = cds[offset:]
return cds
