def initialize_pool_sequence_mappings(self, mapq_cutoff=30):
if self.get_property(
'force_recount'
) or not self.lib_settings.sequence_counts_exist():
gene_names = []
trimmed_sequences = bzUtils.convertFastaToDict(
self.experiment_settings.get_trimmed_pool_fasta())
for sequence_name in trimmed_sequences:
gene_name = sequence_name.split('_')[
0] #TL names are assumed to be of type:YLR350W_-68_651_116
gene_names.append(gene_name)
self.pool_sequence_mappings[
sequence_name] = pool_sequence_mapping(
sequence_name, trimmed_sequences[sequence_name])
samfile = pysam.Samfile(self.lib_settings.get_mapped_reads(), "rb")
ra = read_assigner(self.pool_sequence_mappings, samfile,
mapq_cutoff)
for aligned_read in samfile.fetch():
ra.assign_read(aligned_read)
samfile.close()
self.compute_lib_fractions()
gene_counts = Counter(gene_names)
for mapping_name in self.pool_sequence_mappings:
if gene_counts[mapping_name.split('_')[0]] == 1:
self.pool_sequence_mappings[mapping_name].is_only_tl = True
else:
assert gene_counts[mapping_name.split('_')[0]] != 0
self.pool_sequence_mappings[
mapping_name].is_only_tl = False
bzUtils.makePickle(self.pool_sequence_mappings,
self.lib_settings.get_sequence_counts())
else:
self.pool_sequence_mappings = bzUtils.unPickle(
self.lib_settings.get_sequence_counts())
def get_collapsed_read_fractions(self, lib_settings):
out_name = os.path.join(
self.experiment_settings.get_rdir(), 'QC', 'collapsed_fracs',
'%(sample_name)s.collapsed_read_fractions.pkl' %
{'sample_name': lib_settings.sample_name})
if not tps_utils.file_exists(
out_name) and not self.experiment_settings.get_property(
'force_recollapse'):
collapsed_reads_file = lib_settings.get_collapsed_reads()
read_counts = []
f = gzip.open(collapsed_reads_file)
for line in f:
if not line.strip() == '' and not line.startswith(
'#'): #ignore empty lines and commented out lines
if line.startswith(
'>'): #> marks the start of a new sequence
num_reads = int(line[1:].strip().split('-')[1])
read_counts.append(num_reads)
else:
continue
f.close()
read_fractions = np.array(read_counts) / float(sum(read_counts))
bzUtils.makePickle(read_fractions, out_name)
else:
read_fractions = bzUtils.unPickle(out_name)
return (lib_settings.sample_name, read_fractions)
def get_collapsed_read_fractions(self, lib_settings):
out_name = os.path.join(
self.experiment_settings.get_rdir(),
"QC",
"collapsed_fracs",
"%(sample_name)s.collapsed_read_fractions.pkl" % {"sample_name": lib_settings.sample_name},
)
if not tps_utils.file_exists(out_name) and not self.experiment_settings.get_property("force_recollapse"):
collapsed_reads_file = lib_settings.get_collapsed_reads()
read_counts = []
f = gzip.open(collapsed_reads_file)
for line in f:
if not line.strip() == "" and not line.startswith("#"): # ignore empty lines and commented out lines
if line.startswith(">"): # > marks the start of a new sequence
num_reads = int(line[1:].strip().split("-")[1])
read_counts.append(num_reads)
else:
continue
f.close()
read_fractions = np.array(read_counts) / float(sum(read_counts))
bzUtils.makePickle(read_fractions, out_name)
else:
read_fractions = bzUtils.unPickle(out_name)
return (lib_settings.sample_name, read_fractions)
def initialize_pool_sequence_mappings(self, mapq_cutoff = 30):
if self.get_property('force_recount') or not self.lib_settings.sequence_counts_exist():
gene_names = []
trimmed_sequences = bzUtils.convertFastaToDict(self.experiment_settings.get_trimmed_pool_fasta())
for sequence_name in trimmed_sequences:
gene_name = sequence_name.split('_')[0] #TL names are assumed to be of type:YLR350W_-68_651_116
gene_names.append(gene_name)
self.pool_sequence_mappings[sequence_name] = pool_sequence_mapping(sequence_name, trimmed_sequences[sequence_name])
samfile = pysam.Samfile(self.lib_settings.get_mapped_reads(), "rb" )
ra = read_assigner(self.pool_sequence_mappings, samfile, mapq_cutoff)
for aligned_read in samfile.fetch():
ra.assign_read(aligned_read)
samfile.close()
self.compute_lib_fractions()
gene_counts = Counter(gene_names)
for mapping_name in self.pool_sequence_mappings:
if gene_counts[mapping_name.split('_')[0]]==1:
self.pool_sequence_mappings[mapping_name].is_only_tl = True
else:
assert gene_counts[mapping_name.split('_')[0]] != 0
self.pool_sequence_mappings[mapping_name].is_only_tl = False
bzUtils.makePickle(self.pool_sequence_mappings, self.lib_settings.get_sequence_counts())
else:
self.pool_sequence_mappings = bzUtils.unPickle(self.lib_settings.get_sequence_counts())