def produce_bowtie2_alignments(reads,
index_prefix,
genome_dir,
score_min,
):
bowtie2_options = {'local': True,
#'report_all': True,
'report_up_to': 10,
'seed_mismatches': 1,
'seed_interval_function': 'C,1,0',
'seed_length': 10,
}
sam_file, mappings = mapping_tools.map_bowtie2(index_prefix,
reads=reads,
custom_binary=True,
score_min=score_min,
yield_mappings=True,
**bowtie2_options)
region_fetcher = genomes.build_region_fetcher(genome_dir, load_references=True)
mapping_groups = utilities.group_by(mappings, lambda m: m.qname)
for qname, group in mapping_groups:
group = sorted(group, key=lambda m: (m.tid, m.pos))
alignments = [mapping_to_alignment(mapping, sam_file, region_fetcher)
for mapping in group if not mapping.is_unmapped]
yield qname, alignments
def produce_bowtie2_alignments_old(reads,
sam_fn,
index_prefix,
genome_dir,
score_min,
):
bowtie2_options = {'local': True,
#'report_all': True,
'report_up_to': 10,
'seed_mismatches': 1,
'seed_interval_function': 'C,1,0',
'seed_length': 10,
#'threads': 12,
}
mapping_tools.map_bowtie2(index_prefix,
None,
None,
sam_fn,
unpaired_Reads=reads,
custom_binary=True,
score_min=score_min,
**bowtie2_options)
sam_file = pysam.Samfile(sam_fn)
region_fetcher = genomes.build_region_fetcher(genome_dir, load_references=True)
mapping_groups = utilities.group_by(sam_file, lambda m: m.qname)
for qname, group in mapping_groups:
alignments = [mapping_to_alignment(mapping, sam_file, region_fetcher)
for mapping in group if not mapping.is_unmapped]
yield qname, alignments
def filter_mappings(self):
num_unmapped = 0
num_entirely_genomic = 0
num_nonunique = 0
num_unique = 0
nongenomic_lengths = Counter()
sam_file = pysam.Samfile(self.file_names['accepted_hits'])
region_fetcher = genomes.build_region_fetcher(self.file_names['genome'],
load_references=True,
sam_file=sam_file,
)
extended_sorter = sam.AlignmentSorter(sam_file.references,
sam_file.lengths,
self.file_names['extended'],
)
filtered_sorter = sam.AlignmentSorter(sam_file.references,
sam_file.lengths,
self.file_names['extended_filtered'],
)
extended_mappings = (trim.extend_polyA_end(mapping, region_fetcher) for mapping in sam_file)
mapping_groups = utilities.group_by(extended_mappings, lambda m: m.qname)
with extended_sorter, filtered_sorter:
for qname, group in mapping_groups:
for m in group:
extended_sorter.write(m)
min_nongenomic_length = min(trim.get_nongenomic_length(m) for m in group)
nongenomic_lengths[min_nongenomic_length] += 1
if min_nongenomic_length == 0:
num_entirely_genomic += 1
continue
nonunique = len(group) > 1 or any(m.mapq < 40 for m in group)
if nonunique:
num_nonunique += 1
continue
num_unique += 1
for m in group:
filtered_sorter.write(m)
self.summary.extend(
[('Mapped with no non-genomic A\'s', num_entirely_genomic),
('Nonunique', num_nonunique),
('Unique', num_unique),
],
)
nongenomic_lengths = utilities.counts_to_array(nongenomic_lengths)
self.write_file('nongenomic_lengths', nongenomic_lengths)
def get_transcripts(all_features, genome_dir):
region_fetcher = genomes.build_region_fetcher(genome_dir, load_references=True)
feature_lists = defaultdict(list)
for feature in all_features:
transcript_name = feature.attribute['transcript_id']
feature_lists[transcript_name].append(feature)
transcripts = [Transcript(name, features, None, region_fetcher)
for name, features in feature_lists.iteritems()]
return transcripts
def get_gff_transcripts(all_features, genome_dir):
region_fetcher = genomes.build_region_fetcher(genome_dir, load_references=True)
genes = []
for feature in all_features:
top_level = feature.parent == None
dubious = feature.attribute.get('orf_classification') == 'Dubious'
has_exon = any('exon' in c.feature for c in feature.descendants)
if top_level and has_exon and not dubious:
gene = GFFTranscript(feature, region_fetcher)
genes.append(gene)
return genes
def get_gff_transcripts(all_features, genome_dir):
region_fetcher = genomes.build_region_fetcher(genome_dir,
load_references=True)
genes = []
for feature in all_features:
top_level = feature.parent == None
dubious = feature.attribute.get('orf_classification') == 'Dubious'
has_exon = any('exon' in c.feature for c in feature.descendants)
if top_level and has_exon and not dubious:
gene = GFFTranscript(feature, region_fetcher)
genes.append(gene)
return genes
def call_3p_peaks():
gtf_fn = '/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcriptome/genes.gtf'
genome_dir = '/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/genome/'
composition_fn = '/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcript_recent_As.hdf5'
output_fn = '/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcript_3p_lengths.txt'
region_fetcher = genomes.build_region_fetcher(genome_dir)
CDSs = gtf.get_CDSs(gtf_fn)
CDS_dict = {t.name: t for t in CDSs}
experiments = build_all_experiments(verbose=False)
three_prime_experiments = [(n, e) for n, e in sorted(experiments['three_p_seq']['three_p_seq'].items())] + \
[(n, e) for n, e in sorted(experiments['three_t_fill_seq']['wilkening_nar'].items()) if '3tfill_ypd_rep1' in n] + \
[(n, e) for n, e in sorted(experiments['TIF_seq']['pelechano_nature'].items()) if n == 'ypd_bio1_lib1' or n == 'ypd_bio1_lib4']
argmaxes = {}
fractions = {}
joints = {}
for name, experiment in three_prime_experiments:
print name
argmaxes[name] = {}
fractions[name] = []
joints[name] = []
fn = experiment.file_names['three_prime_read_positions']
f = h5py.File(fn, 'r')
for transcript in utilities.progress_bar(len(CDSs), CDSs):
if transcript.name not in f:
continue
gene = Serialize.read_positions.build_gene(f[transcript.name])
xs = np.arange(0, 400)
argmax = gene['all'].argmax_over_slice('stop_codon', xs)
argmaxes[name][transcript.name] = argmax
most = gene['all']['stop_codon', argmax]
total = gene['all']['stop_codon', xs].sum()
if total > 9:
fraction = np.true_divide(most, total)
fractions[name].append(fraction)
joints[name].append((argmax, fraction))
with open(output_fn, 'w') as output_fh:
name_order = sorted(argmaxes['Cerevisiae_3Pseq'],
key=argmaxes['Cerevisiae_3Pseq'].get)
for name in name_order:
output_fh.write('{0}\t'.format(str(CDS_dict[name])))
for exp_name, _ in three_prime_experiments:
output_fh.write('{0}\t'.format(argmaxes[exp_name][name]))
output_fh.write('\n')
def get_transcripts(all_features, genome_dir):
region_fetcher = genomes.build_region_fetcher(genome_dir,
load_references=True)
feature_lists = defaultdict(list)
for feature in all_features:
transcript_name = feature.attribute['transcript_id']
feature_lists[transcript_name].append(feature)
transcripts = [
Transcript(name, features, None, region_fetcher)
for name, features in feature_lists.iteritems()
]
return transcripts
def get_reads(self):
CDSs, _ = self.get_CDSs()
region_fetcher = genomes.build_region_fetcher(self.file_names['genome'],
load_references=True,
)
for transcript in CDSs:
reads = make_artificial_reads(transcript,
self.fragment_length,
self.max_read_length,
self.adapter_sequence,
region_fetcher,
self.common_buffer,
)
for read in reads:
yield read
def get_reads(self):
CDSs, _ = self.get_CDSs()
region_fetcher = genomes.build_region_fetcher(
self.file_names['genome'],
load_references=True,
)
for transcript in CDSs:
reads = make_artificial_reads(
transcript,
self.fragment_length,
self.max_read_length,
self.adapter_sequence,
region_fetcher,
self.common_buffer,
)
for read in reads:
yield read
def process_full_length_mappings(self):
clean_bam = pysam.Samfile(self.file_names['clean_bam'])
type_shape = (self.max_read_length + 1,
self.max_read_length,
fastq.MAX_EXPECTED_QUAL + 1,
6,
6,
)
type_counts = np.zeros(type_shape, int)
# To avoid counting mismatches in non-unique mappings multiple times,
# a dummy secondary_type_counts array is passed to
# trim_mismatches_from_start for secondary mappings.
secondary_type_counts = np.zeros(type_shape, int)
clean_trimmed_length_counts = Counter()
region_fetcher = genomes.build_region_fetcher(self.file_names['genome'],
load_references=True,
sam_file=clean_bam,
)
for mapping in clean_bam:
if mapping.is_secondary:
counts_array = secondary_type_counts
else:
counts_array = type_counts
trimmed_from_start = trim.trim_mismatches_from_start(mapping,
region_fetcher,
counts_array,
)
trimmed_from_end = trim.trim_nongenomic_polyA_from_end(trimmed_from_start,
region_fetcher,
)
if not trimmed_from_end.is_unmapped and not trimmed_from_end.is_secondary:
clean_trimmed_length_counts[trimmed_from_end.qlen] += 1
yield trimmed_from_end
self.write_file('mismatches', type_counts)
clean_trimmed_lengths = self.zero_padded_array(clean_trimmed_length_counts)
self.write_file('lengths', {'clean_trimmed': clean_trimmed_lengths})
def extend_polyA_ends(bam_fn, extended_bam_fn, genome_dir, trimmed_twice=False):
bam_file = pysam.Samfile(bam_fn)
region_fetcher = genomes.build_region_fetcher(genome_dir,
load_references=True,
sam_file=bam_file,
)
# Adding bases to the end of minus strand mappings produces a file
# that is not necessarily sorted, so re-sort.
alignment_sorter = sam.AlignmentSorter(bam_file.references,
bam_file.lengths,
extended_bam_fn,
)
with alignment_sorter:
for mapping in bam_file:
extended_mapping = extend_polyA_end(mapping, region_fetcher, trimmed_twice)
alignment_sorter.write(extended_mapping)
def call_5p_peaks():
gtf_fn = "/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcriptome/genes.gtf"
genome_dir = "/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/genome/"
region_fetcher = genomes.build_region_fetcher(genome_dir)
CDSs = gtf.get_CDSs(gtf_fn)
experiments = build_all_experiments(verbose=False)
five_prime_experiments = (
[(n, e) for n, e in sorted(experiments["TL_seq"]["arribere_gr"].items()) if "TLSeq1" in n]
+ [(n, e) for n, e in sorted(experiments["TL_seq"]["park_nar"].items()) if n == "SMORE-seq_WT_TAP+_rep1"]
+ [
(n, e)
for n, e in sorted(experiments["TIF_seq"]["pelechano_nature"].items())
if n == "ypd_bio1_lib1" or n == "ypd_bio1_lib4"
]
)
argmaxes = {}
fractions = {}
joints = {}
for name, experiment in five_prime_experiments:
print name
argmaxes[name] = Counter()
fractions[name] = []
joints[name] = []
fn = experiment.file_names["five_prime_read_positions"]
f = h5py.File(fn, "r")
for transcript in utilities.progress_bar(len(CDSs), CDSs):
if transcript.name not in f:
continue
gene = Serialize.read_positions.build_gene(f[transcript.name])
xs = np.arange(-300, 0)
argmax = gene["all"].argmax_over_slice("start_codon", xs)
argmaxes[name][argmax] += 1
most = gene["all"]["start_codon", argmax]
total = gene["all"]["start_codon", xs].sum()
if total == 0:
print transcript
if total > 9:
fraction = np.true_divide(most, total)
fractions[name].append(fraction)
joints[name].append((argmax, fraction))
def compute_yield(self):
lengths = self.read_file('lengths')
reads = {name: lengths[name].sum() for name in lengths}
reads['total'] = reads['trimmed'] + reads['too_short']
reads['dominant'], overlapping_reads, boundaries = contaminants.identify_dominant_stretches(self.read_file('rRNA_coverage'),
reads['total'],
self.max_read_length,
self.merged_file_names['rRNA_bam'],
)
contaminants.plot_dominant_stretch_lengths(boundaries, self.figure_file_names['dominant_stretch_lengths'])
reads['other'] = reads['rRNA'] - reads['dominant']
region_fetcher = genomes.build_region_fetcher(self.file_names['rRNA_fasta_dir'])
with open(self.merged_file_names['dominant_stretches'], 'w') as dominant_stretches_file:
for rname in sorted(boundaries):
for start, stop in sorted(boundaries[rname]):
sequence = region_fetcher(rname, start, stop)
fraction = overlapping_reads[rname, start, stop] / float(reads['total'])
dominant_stretches_file.write('{0}: {1:,}-{2:,}\t{3:6.1%}\t{4}\n'.format(rname, start, stop, fraction, sequence))
with open(self.file_names['yield'], 'w') as yield_file:
yield_file.write('Total reads: {0:,}\n'.format(reads['total']))
for category, count in [('Long enough reads', reads['trimmed']),
('phiX reads', reads['phiX']),
('rRNA reads', reads['rRNA']),
('(rRNA reads from non-dominant stetches)', reads['other']),
('tRNA reads', reads['tRNA']),
('Other ncRNA reads', reads['other_ncRNA']),
('Clean reads', reads['clean']),
('Reads mapped after polyA trimming', reads['remapped']),
('Reads that start wth long polyA', reads['long_polyA']),
('Synthetic reads', reads['synthetic']),
('Unaccounted-for reads', reads['unmapped']),
]:
fraction = float(count) / reads['total']
line = '{0}: {1:,} ({2:.2%})\n'.format(category,
count,
fraction,
)
yield_file.write(line)
def call_5p_peaks():
gtf_fn = '/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcriptome/genes.gtf'
genome_dir = '/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/genome/'
region_fetcher = genomes.build_region_fetcher(genome_dir)
CDSs = gtf.get_CDSs(gtf_fn)
experiments = build_all_experiments(verbose=False)
five_prime_experiments = [(n, e) for n, e in sorted(experiments['TL_seq']['arribere_gr'].items()) if 'TLSeq1' in n] + \
[(n, e) for n, e in sorted(experiments['TL_seq']['park_nar'].items()) if n == 'SMORE-seq_WT_TAP+_rep1'] + \
[(n, e) for n, e in sorted(experiments['TIF_seq']['pelechano_nature'].items()) if n == 'ypd_bio1_lib1' or n == 'ypd_bio1_lib4']
argmaxes = {}
fractions = {}
joints = {}
for name, experiment in five_prime_experiments:
print name
argmaxes[name] = Counter()
fractions[name] = []
joints[name] = []
fn = experiment.file_names['five_prime_read_positions']
f = h5py.File(fn, 'r')
for transcript in utilities.progress_bar(len(CDSs), CDSs):
if transcript.name not in f:
continue
gene = Serialize.read_positions.build_gene(f[transcript.name])
xs = np.arange(-300, 0)
argmax = gene['all'].argmax_over_slice('start_codon', xs)
argmaxes[name][argmax] += 1
most = gene['all']['start_codon', argmax]
total = gene['all']['start_codon', xs].sum()
if total == 0:
print transcript
if total > 9:
fraction = np.true_divide(most, total)
fractions[name].append(fraction)
joints[name].append((argmax, fraction))
def process_remapped(self):
clean_bam = pysam.Samfile(self.file_names['remapped_accepted_hits'])
type_shape = (self.max_read_length + 1,
self.max_read_length,
fastq.MAX_EXPECTED_QUAL + 1,
6,
6,
)
type_counts = np.zeros(type_shape, int)
remapped_length_counts = Counter()
region_fetcher = genomes.build_region_fetcher(self.file_names['genome'],
load_references=True,
sam_file=clean_bam,
)
for mapping in clean_bam:
trimmed_from_start = trim.trim_mismatches_from_start(mapping,
region_fetcher,
type_counts,
)
# Add back any genomic A's that were trimmed as part of mappings and
# any remaining A's from the first non-genomic onward as soft clipped
# bases for visualization in IGV.
extended = trim.extend_polyA_end(trimmed_from_start,
region_fetcher,
trimmed_twice=True,
)
if not extended.is_unmapped and not extended.is_secondary:
remapped_length_counts[extended.qlen] += 1
yield extended
remapped_lengths = self.zero_padded_array(remapped_length_counts)
self.write_file('lengths', {'remapped': remapped_lengths})
def extend_polyA_ends(bam_fn,
extended_bam_fn,
genome_dir,
trimmed_twice=False):
bam_file = pysam.Samfile(bam_fn)
region_fetcher = genomes.build_region_fetcher(
genome_dir,
load_references=True,
sam_file=bam_file,
)
# Adding bases to the end of minus strand mappings produces a file
# that is not necessarily sorted, so re-sort.
alignment_sorter = sam.AlignmentSorter(
bam_file.references,
bam_file.lengths,
extended_bam_fn,
)
with alignment_sorter:
for mapping in bam_file:
extended_mapping = extend_polyA_end(mapping, region_fetcher,
trimmed_twice)
alignment_sorter.write(extended_mapping)
def filter_mappings(self):
num_unmapped = 0
num_entirely_genomic = 0
num_nonunique = 0
num_unique = 0
nongenomic_lengths = Counter()
sam_file = pysam.Samfile(self.file_names['accepted_hits'])
region_fetcher = genomes.build_region_fetcher(
self.file_names['genome'],
load_references=True,
sam_file=sam_file,
)
extended_sorter = sam.AlignmentSorter(
sam_file.references,
sam_file.lengths,
self.file_names['extended'],
)
filtered_sorter = sam.AlignmentSorter(
sam_file.references,
sam_file.lengths,
self.file_names['extended_filtered'],
)
extended_mappings = (trim.extend_polyA_end(mapping, region_fetcher)
for mapping in sam_file)
mapping_groups = utilities.group_by(extended_mappings,
lambda m: m.qname)
with extended_sorter, filtered_sorter:
for qname, group in mapping_groups:
for m in group:
extended_sorter.write(m)
min_nongenomic_length = min(
trim.get_nongenomic_length(m) for m in group)
nongenomic_lengths[min_nongenomic_length] += 1
if min_nongenomic_length == 0:
num_entirely_genomic += 1
continue
nonunique = len(group) > 1 or any(m.mapq < 40 for m in group)
if nonunique:
num_nonunique += 1
continue
num_unique += 1
for m in group:
filtered_sorter.write(m)
self.summary.extend([
('Mapped with no non-genomic A\'s', num_entirely_genomic),
('Nonunique', num_nonunique),
('Unique', num_unique),
], )
nongenomic_lengths = utilities.counts_to_array(nongenomic_lengths)
self.write_file('nongenomic_lengths', nongenomic_lengths)
def call_3p_peaks():
gtf_fn = "/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcriptome/genes.gtf"
genome_dir = "/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/genome/"
composition_fn = (
"/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcript_recent_As.hdf5"
)
output_fn = "/home/jah/projects/ribosomes/data/organisms/saccharomyces_cerevisiae/EF4/transcript_3p_lengths.txt"
region_fetcher = genomes.build_region_fetcher(genome_dir)
CDSs = gtf.get_CDSs(gtf_fn)
CDS_dict = {t.name: t for t in CDSs}
experiments = build_all_experiments(verbose=False)
three_prime_experiments = (
[(n, e) for n, e in sorted(experiments["three_p_seq"]["three_p_seq"].items())]
+ [
(n, e)
for n, e in sorted(experiments["three_t_fill_seq"]["wilkening_nar"].items())
if "3tfill_ypd_rep1" in n
]
+ [
(n, e)
for n, e in sorted(experiments["TIF_seq"]["pelechano_nature"].items())
if n == "ypd_bio1_lib1" or n == "ypd_bio1_lib4"
]
)
argmaxes = {}
fractions = {}
joints = {}
for name, experiment in three_prime_experiments:
print name
argmaxes[name] = {}
fractions[name] = []
joints[name] = []
fn = experiment.file_names["three_prime_read_positions"]
f = h5py.File(fn, "r")
for transcript in utilities.progress_bar(len(CDSs), CDSs):
if transcript.name not in f:
continue
gene = Serialize.read_positions.build_gene(f[transcript.name])
xs = np.arange(0, 400)
argmax = gene["all"].argmax_over_slice("stop_codon", xs)
argmaxes[name][transcript.name] = argmax
most = gene["all"]["stop_codon", argmax]
total = gene["all"]["stop_codon", xs].sum()
if total > 9:
fraction = np.true_divide(most, total)
fractions[name].append(fraction)
joints[name].append((argmax, fraction))
with open(output_fn, "w") as output_fh:
name_order = sorted(argmaxes["Cerevisiae_3Pseq"], key=argmaxes["Cerevisiae_3Pseq"].get)
for name in name_order:
output_fh.write("{0}\t".format(str(CDS_dict[name])))
for exp_name, _ in three_prime_experiments:
output_fh.write("{0}\t".format(argmaxes[exp_name][name]))
output_fh.write("\n")
def combine_mappings(self):
num_unmapped = 0
num_five_unmapped = 0
num_three_unmapped = 0
num_nonunique = 0
num_discordant = 0
num_concordant = 0
five_prime_mappings = pysam.Samfile(self.file_names['five_prime_accepted_hits'])
five_prime_unmapped = pysam.Samfile(self.file_names['five_prime_unmapped'])
all_five_prime = sam.merge_by_name(five_prime_mappings, five_prime_unmapped)
five_prime_grouped = utilities.group_by(all_five_prime, lambda m: m.qname)
three_prime_mappings = pysam.Samfile(self.file_names['three_prime_accepted_hits'])
three_prime_unmapped = pysam.Samfile(self.file_names['three_prime_unmapped'])
all_three_prime = sam.merge_by_name(three_prime_mappings, three_prime_unmapped)
three_prime_grouped = utilities.group_by(all_three_prime, lambda m: m.qname)
group_pairs = izip(five_prime_grouped, three_prime_grouped)
alignment_sorter = sam.AlignmentSorter(five_prime_mappings.references,
five_prime_mappings.lengths,
self.file_names['combined_extended'],
)
region_fetcher = genomes.build_region_fetcher(self.file_names['genome'],
load_references=True,
sam_file=five_prime_mappings,
)
with alignment_sorter:
for (five_qname, five_group), (three_qname, three_group) in group_pairs:
five_annotation = trim.PayloadAnnotation.from_identifier(five_qname)
three_annotation = trim.PayloadAnnotation.from_identifier(three_qname)
if five_annotation['original_name'] != three_annotation['original_name']:
# Ensure that the iteration through pairs is in sync.
print five_qname, three_qname
raise ValueError
five_unmapped = any(m.is_unmapped for m in five_group)
three_unmapped = any(m.is_unmapped for m in three_group)
if five_unmapped:
num_five_unmapped += 1
if three_unmapped:
num_three_unmapped += 1
if five_unmapped or three_unmapped:
num_unmapped += 1
continue
five_nonunique = len(five_group) > 1 or any(m.mapq < 40 for m in five_group)
three_nonunique = len(three_group) > 1 or any(m.mapq < 40 for m in three_group)
if five_nonunique or three_nonunique:
num_nonunique += 1
continue
five_m = five_group.pop()
three_m = three_group.pop()
five_strand = '-' if five_m.is_reverse else '+'
three_strand = '-' if three_m.is_reverse else '+'
tlen = max(five_m.aend, three_m.aend) - min(five_m.pos, three_m.pos)
discordant = (five_m.tid != three_m.tid) or (five_strand) != (three_strand) or (tlen > 10000)
if discordant:
num_discordant += 1
continue
if five_strand == '+':
first_read = five_m
second_read = three_m
elif five_strand == '-':
first_read = three_m
second_read = five_m
gap = second_read.pos - first_read.aend
if gap < 0:
num_discordant += 1
continue
combined_read = pysam.AlignedRead()
# qname needs to come from three_m to include trimmed As
combined_read.qname = three_m.qname
combined_read.tid = five_m.tid
combined_read.seq = first_read.seq + second_read.seq
combined_read.qual = first_read.qual + second_read.qual
combined_read.cigar = first_read.cigar + [(3, gap)] + second_read.cigar
combined_read.pos = first_read.pos
combined_read.is_reverse = first_read.is_reverse
combined_read.mapq = min(first_read.mapq, second_read.mapq)
combined_read.rnext = -1
combined_read.pnext = -1
num_concordant += 1
extended_mapping = trim.extend_polyA_end(combined_read,
region_fetcher,
)
alignment_sorter.write(extended_mapping)
self.summary.extend(
[('Unmapped', num_unmapped),
('Five prime unmapped', num_five_unmapped),
('Three prime unmapped', num_three_unmapped),
('Nonunique', num_nonunique),
('Discordant', num_discordant),
('Concordant', num_concordant),
],
)
