def __init__(self, parent, base_dir, lower_bound, upper_bound):
# Save parent #
self.parent, self.fractions = parent, parent
# Auto paths #
self.base_dir = base_dir
self.p = AutoPaths(self.base_dir, self.all_paths)
# Bounds #
self.lower_bound = lower_bound
self.upper_bound = upper_bound
# Size fractions #
self.reads = FASTA(self.p.reads_fasta)
self.refere = UchimeRef(self.p.reads, self.p.refere_dir, self)
self.denovo = UchimeDenovo(self.p.reads, self.p.denovo_dir, self)
# Classification #
self.rdp = SimpleRdpTaxonomy(self.reads, self.p.rdp_dir)
self.crest = SimpleCrestTaxonomy(self.reads, self.p.crest_dir)
class Fraction(object):
"""One size fraction from the quality reads."""
all_paths = """
/reads.fasta
/refere/
/denovo/
/rdp/
/crest/
"""
def __init__(self, parent, base_dir, lower_bound, upper_bound):
# Save parent #
self.parent, self.fractions = parent, parent
# Auto paths #
self.base_dir = base_dir
self.p = AutoPaths(self.base_dir, self.all_paths)
# Bounds #
self.lower_bound = lower_bound
self.upper_bound = upper_bound
# Size fractions #
self.reads = FASTA(self.p.reads_fasta)
self.refere = UchimeRef(self.p.reads, self.p.refere_dir, self)
self.denovo = UchimeDenovo(self.p.reads, self.p.denovo_dir, self)
# Classification #
self.rdp = SimpleRdpTaxonomy(self.reads, self.p.rdp_dir)
self.crest = SimpleCrestTaxonomy(self.reads, self.p.crest_dir)
def extract(self):
self.fractions.pool.quality_reads.only_used.extract_length(self.lower_bound, self.upper_bound, self.reads)
return self.reads
def check_chimeras(self):
self.refere.check()
self.denovo.check()
return self.refere, self.denovo
def check_classification(self):
self.rdp.assign()
class Fraction(object):
"""One size fraction from the quality reads."""
all_paths = """
/reads.fasta
/refere/
/denovo/
/rdp/
/crest/
"""
def __init__(self, parent, base_dir, lower_bound, upper_bound):
# Save parent #
self.parent, self.fractions = parent, parent
# Auto paths #
self.base_dir = base_dir
self.p = AutoPaths(self.base_dir, self.all_paths)
# Bounds #
self.lower_bound = lower_bound
self.upper_bound = upper_bound
# Size fractions #
self.reads = FASTA(self.p.reads_fasta)
self.refere = UchimeRef(self.p.reads, self.p.refere_dir, self)
self.denovo = UchimeDenovo(self.p.reads, self.p.denovo_dir, self)
# Classification #
self.rdp = SimpleRdpTaxonomy(self.reads, self.p.rdp_dir)
self.crest = SimpleCrestTaxonomy(self.reads, self.p.crest_dir)
def extract(self):
self.fractions.pool.quality_reads.only_used.extract_length(self.lower_bound, self.upper_bound, self.reads)
return self.reads
def check_chimeras(self):
self.refere.check()
self.denovo.check()
return self.refere, self.denovo
def check_classification(self):
self.rdp.assign()
def __init__(self, parent, base_dir, lower_bound, upper_bound):
# Save parent #
self.parent, self.fractions = parent, parent
# Auto paths #
self.base_dir = base_dir
self.p = AutoPaths(self.base_dir, self.all_paths)
# Bounds #
self.lower_bound = lower_bound
self.upper_bound = upper_bound
# Size fractions #
self.reads = FASTA(self.p.reads_fasta)
self.refere = UchimeRef(self.p.reads, self.p.refere_dir, self)
self.denovo = UchimeDenovo(self.p.reads, self.p.denovo_dir, self)
# Classification #
self.rdp = SimpleRdpTaxonomy(self.reads, self.p.rdp_dir)
self.crest = SimpleCrestTaxonomy(self.reads, self.p.crest_dir)
# Make fraction graph #
proj.graphs[-1].plot()
# Get statistics #
proj.reporter.fraction_discarded
# Get clustering values #
r1, r2 = list(set([p.run for p in proj]))
r1.parse_report_xml()
r2.parse_report_xml()
print float(r1.report_stats['fwd']['DensityPF']) / float(r1.report_stats['fwd']['DensityRaw'])
print float(r2.report_stats['fwd']['DensityPF']) / float(r2.report_stats['fwd']['DensityRaw'])
# Check below 400 bp sequences #
folder = DirectoryPath(illumitag.projects['evaluation'].base_dir + "below_400/")
over = FASTA(folder + "reads.fasta")
def over_iterator(reads, max_length=400):
for read in reads:
if len(read) <= max_length: yield read
over.create()
for pool in pools: over.add_iterator(over_iterator(pool.good_barcodes.assembled.good_primers.qual_filtered))
over.close()
over.graphs[-1].plot()
crest = SimpleCrestTaxonomy(over, folder)
crest.assign()
crest.composition.graph.plot()
rdp = SimpleRdpTaxonomy(over, folder)
rdp.assign()
rdp.composition.graph.plot()
print float(r2.report_stats['fwd']['DensityPF']) / float(r2.report_stats['fwd']['DensityRaw'])
# Check below 400 bp sequences #
folder = DirectoryPath(illumitag.projects['evaluation'].base_dir + "below_400/")
over = FASTA(folder + "reads.fasta")
def over_iterator(reads, max_length=400):
for read in reads:
if len(read) <= max_length: yield read
over.create()
for pool in pools: over.add_iterator(over_iterator(pool.good_barcodes.assembled.good_primers.qual_filtered))
over.close()
over.graphs[-1].plot()
crest = SimpleCrestTaxonomy(over, folder)
crest.assign()
crest.composition.graph.plot()
rdp = SimpleRdpTaxonomy(over, folder)
rdp.assign()
rdp.composition.graph.plot()
# Check unassembled mate pairs #
unassembled = [p.good_barcodes.unassembled for p in pools]
paths = [u.flipped_reads.path for u in unassembled]
folder = DirectoryPath(illumitag.projects['evaluation'].base_dir + "unassembled_taxonomy/")
all_unassembled = FASTA(folder + 'unassembled_reads.fasta')
shell_output('cat %s > %s' % (' '.join(paths), all_unassembled))
tax = SimpleRdpTaxonomy(all_unassembled, folder)
tax.assign()
tax.composition.graph.plot()
# Upload raw samples for ENA #
for pool in pools: pool.create_raw_samples()