def count_kmers(self):
# We need to clear all previous primers each time due to uniqueness
# constraints
if Primer.select().count() > 0:
if not self.force:
click.confirm(
"Remove all previously-found primers and re-count?",
abort=True)
self.workspace.reset_primers()
mkdirp(output_dir)
kmers = []
for k in xrange(self.min_size, self.max_size + 1):
fg = swga.kmers.count_kmers(k, self.fg_genome_fp, output_dir)
bg = swga.kmers.count_kmers(k, self.bg_genome_fp, output_dir)
if self.exclude_fp:
assert os.path.isfile(self.exclude_fp)
ex = swga.kmers.count_kmers(k, self.exclude_fp, output_dir,
self.exclude_threshold)
else:
ex = {}
# Keep kmers found in foreground, merging bg binding values, and
# excluding those found in the excluded fasta
kmers = [
primer_dict(seq, fg, bg, self.min_fg_bind, self.max_bg_bind,
self.max_dimer_bp) for seq in fg.viewkeys()
if seq not in ex.viewkeys()
]
kmers = filter(lambda x: x != {}, kmers)
nkmers = len(kmers)
chunk_size = 199
message("Writing {n} {k}-mers into db in blocks of {cs}...".format(
n=nkmers * 2, k=k, cs=chunk_size))
Primers.add(kmers, add_revcomp=True)
message("Counted kmers in range %d-%d" %
(self.min_size, self.max_size))
def count_kmers(self):
# We need to clear all previous primers each time due to uniqueness
# constraints
if Primer.select().count() > 0:
if not self.force:
click.confirm(
"Remove all previously-found primers and re-count?",
abort=True)
self.workspace.reset_primers()
mkdirp(output_dir)
kmers = []
for k in xrange(self.min_size, self.max_size + 1):
fg = swga.kmers.count_kmers(k, self.fg_genome_fp, output_dir)
bg = swga.kmers.count_kmers(k, self.bg_genome_fp, output_dir)
if self.exclude_fp:
assert os.path.isfile(self.exclude_fp)
ex = swga.kmers.count_kmers(
k, self.exclude_fp, output_dir, self.exclude_threshold)
else:
ex = {}
# Keep kmers found in foreground, merging bg binding values, and
# excluding those found in the excluded fasta
kmers = [
primer_dict(seq, fg, bg, self.min_fg_bind, self.max_bg_bind,
self.max_dimer_bp)
for seq in fg.viewkeys() if seq not in ex.viewkeys()
]
kmers = filter(lambda x: x != {}, kmers)
nkmers = len(kmers)
chunk_size = 199
message(
"Writing {n} {k}-mers into db in blocks of {cs}..."
.format(n=nkmers * 2, k=k, cs=chunk_size))
Primers.add(kmers, add_revcomp=True)
message("Counted kmers in range %d-%d" % (self.min_size, self.max_size))
def count_specific_kmers(self, kmers):
try:
# Skip primers that already exist and warn users
existing = Primers.select_by_seqs(kmers)
for p in existing:
message("{} already exists in db, skipping...".format(p))
kmers = [p for p in kmers if p not in existing]
except OperationalError:
# If this fails due to an OperationalError, it probably means the
# database tables haven't been created yet.
error(
"It doesn't appear that the workspace has been initialized: "
"run `swga init' first.")
mkdirp(output_dir)
# Group the kmers by length to avoid repeatedly counting kmers of the
# same size
kmers_by_length = defaultdict(list)
for kmer in kmers:
kmers_by_length[len(kmer)].append(kmer)
for k, mers in kmers_by_length.items():
fg = swga.kmers.count_kmers(k, self.fg_genome_fp, output_dir, 1)
bg = swga.kmers.count_kmers(k, self.bg_genome_fp, output_dir, 1)
primers = []
for mer in mers:
try:
primers.append(primer_dict(mer, fg, bg, 0, INF, INF))
except KeyError:
message(
"{} does not exist in foreground genome, skipping..."
.format(mer))
# Omitting any primers that were returned empty
# primers = filter(lambda p: p == {}, primers)
chunk_size = 199
message(
"Writing {n} {k}-mers into db in blocks of {cs}..."
.format(n=len(primers), k=k, cs=chunk_size))
Primers.add(primers, add_revcomp=False)
def count_specific_kmers(self, kmers):
try:
# Skip primers that already exist and warn users
existing = Primers.select_by_seqs(kmers)
for p in existing:
message("{} already exists in db, skipping...".format(p))
kmers = [p for p in kmers if p not in existing]
except OperationalError:
# If this fails due to an OperationalError, it probably means the
# database tables haven't been created yet.
error("It doesn't appear that the workspace has been initialized: "
"run `swga init' first.")
mkdirp(output_dir)
# Group the kmers by length to avoid repeatedly counting kmers of the
# same size
kmers_by_length = defaultdict(list)
for kmer in kmers:
kmers_by_length[len(kmer)].append(kmer)
for k, mers in kmers_by_length.items():
fg = swga.kmers.count_kmers(k, self.fg_genome_fp, output_dir, 1)
bg = swga.kmers.count_kmers(k, self.bg_genome_fp, output_dir, 1)
primers = []
for mer in mers:
try:
primers.append(primer_dict(mer, fg, bg, 0, INF, INF))
except KeyError:
message(
"{} does not exist in foreground genome, skipping...".
format(mer))
# Omitting any primers that were returned empty
# primers = filter(lambda p: p == {}, primers)
chunk_size = 199
message("Writing {n} {k}-mers into db in blocks of {cs}...".format(
n=len(primers), k=k, cs=chunk_size))
Primers.add(primers, add_revcomp=False)
def test_add_primers(self):
'''Must add the reverse complement of a primer if requested.'''
primers = [{'seq': "AAAA"}]
Primers.add(primers, add_revcomp=True)
assert Primer.select().where(Primer.seq == "TTTT").count() == 1