class QualityReads(object):
"""A set of sequences determined to be quality controlled"""
all_paths = """
/mothur_reads.fasta
/mothur_reads.qual
/mothur_groups.tsv
/qiime_reads.fasta
/only_used_samples.fasta
/trimmed.fasta
"""
def __repr__(self): return '<%s object of %s>' % (self.__class__.__name__, self.parent)
def __len__(self): return len(self.trimmed)
def __init__(self, path, parent):
# Save parent #
self.parent, self.pool = parent, parent
self.samples = parent.samples
# Auto paths #
self.base_dir = parent.p.quality_dir + '/'
self.p = AutoPaths(self.base_dir, self.all_paths)
# Files #
self.untrimmed = BarcodedFASTQ(path, samples=self.samples)
self.only_used = BarcodedFASTA(self.p.only_used, samples=self.samples)
self.trimmed = FASTA(self.p.trimmed)
# Qiime output #
self.qiime_fasta = FASTA(self.p.qiime_fasta)
# Mothur #
self.mothur_fasta = FASTA(self.p.mothur_fasta)
self.mothur_qual = QualFile(self.p.mothur_qual)
self.mothur_groups = FilePath(self.p.mothur_groups)
# Primer size #
self.trim_fwd = self.pool.samples.trim_fwd
self.trim_rev = self.pool.samples.trim_rev
def filter_unused(self):
def no_unused_iterator(reads):
for r in reads.parse_barcodes():
if r.first.sample.used: yield r.read
self.only_used.write(no_unused_iterator(self.untrimmed))
def trim_primers(self):
def no_primers_iterator(reads):
for read in reads:
yield read[self.trim_fwd:-self.trim_rev]
self.trimmed.write(no_primers_iterator(self.only_used))
def make_mothur_output(self):
# Trimmed fasta #
self.mothur_fasta.link_from(self.trimmed.path)
# The groups file #
self.mothur_groups.create()
for r in self.only_used.parse_barcodes():
sample_name = r.first.sample.short_name
read_name = '%s\t%s\n' % (r.read.id, sample_name)
self.mothur_groups.handle.write(read_name)
self.mothur_groups.close()
def make_qiime_output(self):
# Prepare fasta writer #
handle = open(self.qiime_fasta.path, 'w')
writer = FastaWriter(handle, wrap=0)
writer.write_header()
# Counter #
counter = defaultdict(int)
# Do it #
for r in self.only_used.parse_barcodes():
sample_name = r.first.sample.short_name
counter[sample_name] += 1
r.read.id = '%s_%i %s' % (sample_name, counter[sample_name], r.read.id)
bar_seq = r.read.seq[0:self.pool.bar_len]
r.read.description = "orig_bc=%s new_bc=%s bc_diffs=0" % (bar_seq, bar_seq)
writer.write_record(r.read[self.trim_fwd:-self.trim_rev])
# Close #
writer.write_footer()
handle.close()
class Foraminifera(Database):
"""This is a custom database containing exlcusively Foraminifera sequences.
https://genev.unige.ch/research/laboratory/Jan-Pawlowski
You should place the file "foram_db_cor.fasta" in: ~/databases/foraminifera/
Then you can run this:
from seqsearch.databases.foraminifera import foraminifera
foraminifera.process()
print foraminifera.tax_depth_freq
"""
short_name = "foraminifera"
long_name = 'The custom made Foraminifera database as received by email on 7th April 2017'
all_paths = """
/foram_db_cor.fasta
/foram_mothur.fasta
/foram_mothur.tax
"""
@property
def rank_names(self):
"""The names of the ranks. Total 9 ranks."""
return ['Domain', # 0
'Kingdom', # 1
'Phylum', # 2
'Class', # 3
'Order', # 4
'Family', # 5
'Tribe', # 6
'Genus', # 7
'Species'] # 8
def __init__(self, base_dir=None):
# Base directory #
if base_dir is None: base_dir = home
self.base_dir = base_dir + 'databases/' + self.short_name + '/'
self.p = AutoPaths(self.base_dir, self.all_paths)
# The results #
self.alignment = FASTA(self.p.mothur_fasta)
self.taxonomy = FilePath(self.p.mothur_tax)
# The part that mothur will use for naming files #
self.nickname = "foram_mothur"
def process(self):
# The file that was received by email without documentation T_T #
raw = FASTA(self.p.cor)
# Open files #
self.alignment.create()
self.taxonomy.create()
# Loop #
for seq in raw:
# Parse #
name = seq.id[11:].split('|')
num = name.pop(0)
# Check #
for x in name: assert ';' not in x
for x in name: assert '\t' not in x
# Make ranks #
ranks = ['Eukaryota' , # 0 Domain
'Rhizaria' , # 1 Kingdom
'Foraminifera' , # 2 Phylum
name[0] , # 3 Class
name[1] , # 4 Order
name[2] , # 5 Family
name[3] , # 6 Tribe
name[4] , # 7 Genus
name[5]] # 8 Species
# The taxonomy string #
tax_line = ';'.join(ranks)
# Add sequence to the new fasta file #
self.alignment.add_str(str(seq.seq), name="foram" + num)
# Add the taxonomy to the tax file #
self.taxonomy.add_str("foram" + num + '\t' + tax_line + '\n')
# Close files #
self.alignment.close()
self.taxonomy.close()