def run(self):
# If we have an input file, use that. Otherwise pull from db
if self.input:
with open(self.input, 'rb') as infile:
primers = Primers(infile)
else:
self.skip_filtering = False
primers = Primers()
assert isinstance(primers, Primers)
# Undo all active marks, if any
Primer.update(active=False).execute()
if not self.skip_filtering:
(
primers
.filter_min_fg_rate(self.min_fg_bind)
.filter_max_bg_rate(self.max_bg_bind)
.summarize()
.filter_tm_range(self.min_tm, self.max_tm)
.limit_to(self.max_primers)
.filter_max_gini(self.max_gini, self.fg_genome_fp)
)
primers.activate(self.max_primers)
def run(self):
primers = Primers(self.input)
try:
(primers.update_melt_temps().update_locations(
self.fg_genome_fp).activate())
except AttributeError as e:
warn("Error updating database: '{}'".format(e.message))
raise e
def run(self):
self.chr_ends = locate.chromosome_ends(self.fg_genome_fp)
# Evaluate the scoring expression from a string and return it as a
# callable function
self.score_fun = functools.partial(score.default_score_set,
expression=self.score_expression)
primers = Primers(self.input)
if len(primers) == 0:
error("No primers specified exist in database, aborting.",
exception=False)
bg_dist_mean = score.calculate_bg_dist_mean(primers, self.bg_length)
set_score, variables, _ = score.score_set(primers=primers,
max_fg_bind_dist=0,
bg_dist_mean=bg_dist_mean,
chr_ends=self.chr_ends,
score_fun=self.score_fun,
interactive=True)
do_add_set, set_id = self.user_add_set(set_score, variables)
if do_add_set:
s = workspace.Set.add(_id=set_id,
primers=primers,
score=set_score,
scoring_fn=self.score_expression,
**variables)
set_added = s is not None
if set_added:
message("Set {} added successfully.".format(set_id))
else:
message("That primer set already exists.")
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 process_lines(self, setfinder_lines):
passed = processed = 0
smallest_max_dist = float('inf')
try:
for line in setfinder_lines:
try:
primer_ids, bg_dist_mean = score.read_set_finder_line(line)
except ValueError:
warn("Could not parse line:\n\t" + line)
continue
primers = Primers.select_by_ids(primer_ids)
processed += 1
set_score, variables, max_dist = score.score_set(
primers=primers,
max_fg_bind_dist=self.max_fg_bind_dist,
bg_dist_mean=bg_dist_mean,
chr_ends=self.chr_ends,
score_fun=self.score_fun,
interactive=False
)
if max_dist < smallest_max_dist:
smallest_max_dist = max_dist
message(
STATUS_LINE.format(processed, passed, smallest_max_dist),
newline=False)
# Return early if the set doesn't pass
if set_score is False:
continue
else:
passed += 1
Set.add(
_id=passed,
primers=primers,
score=set_score,
scoring_fn=self.score_expression,
**variables)
if passed >= self.max_sets:
message("\nDone (scored %i sets)" % passed)
break
finally:
# Raises a GeneratorExit inside the find_sets command, prompting it
# to quit the subprocess
setfinder_lines.close()
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 process_lines(self, setfinder_lines):
passed = processed = 0
smallest_max_dist = float('inf')
try:
for line in setfinder_lines:
try:
primer_ids, bg_dist_mean = score.read_set_finder_line(line)
except ValueError:
warn("Could not parse line:\n\t" + line)
continue
primers = Primers.select_by_ids(primer_ids)
processed += 1
set_score, variables, max_dist = score.score_set(
primers=primers,
max_fg_bind_dist=self.max_fg_bind_dist,
bg_dist_mean=bg_dist_mean,
chr_ends=self.chr_ends,
score_fun=self.score_fun,
interactive=False)
if max_dist < smallest_max_dist:
smallest_max_dist = max_dist
message(STATUS_LINE.format(processed, passed,
smallest_max_dist),
newline=False)
# Return early if the set doesn't pass
if set_score is False:
continue
else:
passed += 1
Set.add(_id=passed,
primers=primers,
score=set_score,
scoring_fn=self.score_expression,
**variables)
if passed >= self.max_sets:
message("\nDone (scored %i sets)" % passed)
break
finally:
# Raises a GeneratorExit inside the find_sets command, prompting it
# to quit the subprocess
setfinder_lines.close()
def build_graph(max_hetdimer_bind, outfile):
'''Selects all active primers and outputs a primer compatibility graph.'''
# Reset all the primer IDs (as ids are only used for set_finder)
primers = Primers.select_active().assign_ids()
# print [(p._id, p.ratio) for p in primers]
message("Composing primer compatibility graph...")
edges = build_edges(primers, max_hetdimer_bind)
if len(edges) == 0:
error("No compatible primers. Try relaxing your parameters.",
exception=False)
with open(outfile, 'wb') as out:
write_graph(primers, edges, out)
def build_graph(max_hetdimer_bind, outfile):
'''Selects all active primers and outputs a primer compatibility graph.'''
# Reset all the primer IDs (as ids are only used for set_finder)
primers = Primers.select_active().assign_ids()
# print [(p._id, p.ratio) for p in primers]
message("Composing primer compatibility graph...")
edges = build_edges(primers, max_hetdimer_bind)
if len(edges) == 0:
error(
"No compatible primers. Try relaxing your parameters.",
exception=False)
with open(outfile, 'wb') as out:
write_graph(primers, edges, out)
def run(self):
# If we have an input file, use that. Otherwise pull from db
if self.input:
with open(self.input, 'rb') as infile:
primers = Primers(infile)
else:
self.skip_filtering = False
primers = Primers()
assert isinstance(primers, Primers)
# Undo all active marks, if any
Primer.update(active=False).execute()
if not self.skip_filtering:
(primers.filter_min_fg_rate(self.min_fg_bind).filter_max_bg_rate(
self.max_bg_bind).summarize().filter_tm_range(
self.min_tm,
self.max_tm).limit_to(self.max_primers).filter_max_gini(
self.max_gini, self.fg_genome_fp))
primers.activate(self.max_primers)
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