def get_reads():
for i, (seq, count) in enumerate(self.read_file('common_unmapped')['non_long_polyA'].most_common()):
read = fastq.Read('{0}_{1}'.format(i, count),
seq,
fastq.encode_sanger([40]*len(seq)),
)
yield read
def make_artificial_reads(
transcript,
fragment_length,
read_length,
adapter_sequence,
region_fetcher,
common_buffer,
):
transcript_sequence = transcript.retrieve_sequence(
region_fetcher,
left_buffer=common_buffer,
right_buffer=common_buffer + fragment_length,
)
# Needs to include one non-Solexa value for automatic encoding recognition.
high_quals = fastq.encode_sanger([25] + [30] * (read_length - 1))
for i, transcript_position in enumerate(
range(-common_buffer, transcript.CDS_length + common_buffer)):
annotation = artifical_annotation(
transcript_name=transcript.name,
position=transcript_position,
)
fragment_sequence = transcript_sequence[i:i + fragment_length]
if '-' in fragment_sequence:
# skip fragments that run off the edge of a reference sequence
continue
full_sequence = fragment_sequence + adapter_sequence
read = fastq.Read(annotation.identifier, full_sequence[:read_length],
high_quals)
yield read
def get_reads():
for i, (seq, count) in enumerate(counts.read_file(unmapped_fn).most_common()):
read = fastq.Read('{0}_{1}'.format(i, count),
seq,
fastq.encode_sanger([40]*len(seq)),
)
yield read
def make_artificial_reads(transcript,
fragment_length,
read_length,
adapter_sequence,
region_fetcher,
common_buffer,
):
transcript_sequence = transcript.retrieve_sequence(region_fetcher,
left_buffer=common_buffer,
right_buffer=common_buffer + fragment_length,
)
# Needs to include one non-Solexa value for automatic encoding recognition.
high_quals = fastq.encode_sanger([25] + [30]*(read_length - 1))
for i, transcript_position in enumerate(range(-common_buffer, transcript.CDS_length + common_buffer)):
annotation = artifical_annotation(transcript_name=transcript.name,
position=transcript_position,
)
fragment_sequence = transcript_sequence[i:i + fragment_length]
if '-' in fragment_sequence:
# skip fragments that run off the edge of a reference sequence
continue
full_sequence = fragment_sequence + adapter_sequence
read = fastq.Read(annotation.identifier, full_sequence[:read_length], high_quals)
yield read
