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 primers():
primers = [
Primer.create(seq="ATGC", fg_freq=1, bg_freq=2, ratio=1.0, active=True),
Primer.create(seq="GGCC", fg_freq=1, bg_freq=3, ratio=0.5, active=True),
Primer.create(seq="CCTA", fg_freq=2, bg_freq=0, ratio=float('inf'), active=True)
]
return primers
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_linearize_binding_sites(kmer, ws, fastafile):
p = Primer.create(seq=kmer)
p._update_locations(fastafile)
chr_ends = swga.locate.chromosome_ends(fastafile)
linear_bind_sites = swga.locate.linearize_binding_sites([p], chr_ends)
# (number of sites + (2*number of chromosomes) - (any overlaps))
assert len(linear_bind_sites) == 10
for record, ends in chr_ends.iteritems():
start, end = ends
assert start in linear_bind_sites
assert end in linear_bind_sites
for site in p.locations[record]:
assert site in linear_bind_sites
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 primers(self):
primers = [
# reference primer
Primer.create(_id=0, seq="ATGCTC"),
# rev. complement has 4 bases overlapping
Primer.create(_id=1, seq="CAGCAT"),
# rev. complement has 3 bases overlapping
Primer.create(_id=2, seq="GAGGTA"),
Primer.create(_id=3, seq="ATCGAG"),
# rev. complement has one base overlapping
Primer.create(_id=4, seq="TTCCAC"),
# substring of reference primer
Primer.create(_id=5, seq="ATGC")
]
return primers
def primers(self):
primers = [
# reference primer
Primer.create(_id=0, seq="ATGCTC"),
# rev. complement has 4 bases overlapping
Primer.create(_id=1, seq="CAGCAT"),
# rev. complement has 3 bases overlapping
Primer.create(_id=2, seq="GAGGTA"),
Primer.create(_id=3, seq="ATCGAG"),
# rev. complement has one base overlapping
Primer.create(_id=4, seq="TTCCAC"),
# substring of reference primer
Primer.create(_id=5, seq="ATGC")
]
return primers
def primers(ws, seqs):
return [Primer.create(seq=seq) for seq in seqs]
def primers(ws, seqs):
return [Primer.create(seq=seq) for seq in seqs]
def run(self):
self.summary_msg = summary_template
self.best_set_desc = best_set_desc
avg_fg_bind, avg_bg_bind, nprimers = (
Primer
.select(fn.Avg(Primer.fg_freq),
fn.Avg(Primer.bg_freq),
fn.Count(Primer.seq))
.scalar(as_tuple=True))
if (avg_fg_bind is None) or (avg_bg_bind is None):
(avg_fg_bind, avg_bg_bind) = (0, 0)
fg_bind_ratio = avg_fg_bind / float(self.fg_length)
bg_bind_ratio = avg_bg_bind / float(self.bg_length)
nactive = Primer.select().where(Primer.active == True).count()
min_tm, max_tm, avg_tm = (
Primer
.select(fn.Min(Primer.tm),
fn.Max(Primer.tm),
fn.Avg(Primer.tm))
.where(Primer.active == True)
.scalar(as_tuple=True))
nsets = Set.select(fn.Count(Set._id)).scalar()
if nsets > 0:
bs = Set.select().order_by(Set.score).limit(1).get()
bs_primers = ", ".join(bs.primer_seqs()).strip()
best_set = bs._id
bs_size = bs.set_size
bs_score = bs.score
bs_stats = "- " + "\n - ".join(
fmtkv(k, bs.__dict__['_data'][k])
for k in bs.exported_fields()
if k not in ["_id", "pids", "score", "primers"]
)
self.best_set_desc = self.best_set_desc.format(**locals())
if_no_primers_msg = click.style(
"Run `swga count` to find possible primers."
if nprimers == 0 else "", fg='green')
if_no_active_primers_msg = click.style(
"Run `swga filter` to identify primers to use."
if nactive == 0 else "", fg='green')
melting_tmp_msg = (
"The melting temp of the primers ranges between {min_tm:.2f}C and "
"{max_tm:.2f}C with an average of {avg_tm:.2f}C."
if nactive > 0 and min_tm and max_tm else
"No melting temps have been calculated yet.").format(**locals())
ifzero_sets_msg = click.style(
"Run `swga find_sets` after identifying valid primers to begin "
"collecting sets.\n", fg='green')
set_msg = (self.best_set_desc if nsets > 0 else ifzero_sets_msg)
primer_db = os.path.abspath(self.primer_db)
nprimers = click.style(str(nprimers), bold=True, fg='blue')
nactive = click.style(str(nactive), bold=True, fg='blue')
nsets = click.style(str(nsets), bold=True, fg='blue')
self.header = click.style("swga v{}".format(__version__), fg='green')
# Copy all the relevant values into one dict
values = self.__dict__.copy()
values.update(locals())
# Format the summary message with all the calculated values
self.summary_msg = self.summary_msg.format(**values)
click.echo(quote(self.summary_msg, quote=" ", width=200))
def validate_order_field(field, model):
'''Ensures the given field exists in the model.'''
if field and field not in model.fields():
swga.error(
"Cannot order by '{}'. Valid choices are {}"
.format(field, ", ".join(Primer.fields())))
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
def test_bad_set_add(self):
with pytest.raises(ValueError):
Set.add(0, None, score=1)
with pytest.raises(ValueError):
invalid_primers = Primer.select().where(Primer.seq == "XX")
Set.add(0, primers=invalid_primers, score=1)
def run(self):
self.summary_msg = summary_template
self.best_set_desc = best_set_desc
avg_fg_bind, avg_bg_bind, nprimers = (Primer.select(
fn.Avg(Primer.fg_freq), fn.Avg(Primer.bg_freq),
fn.Count(Primer.seq)).scalar(as_tuple=True))
if (avg_fg_bind is None) or (avg_bg_bind is None):
(avg_fg_bind, avg_bg_bind) = (0, 0)
fg_bind_ratio = avg_fg_bind / float(self.fg_length)
bg_bind_ratio = avg_bg_bind / float(self.bg_length)
nactive = Primer.select().where(Primer.active == True).count()
min_tm, max_tm, avg_tm = (Primer.select(fn.Min(
Primer.tm), fn.Max(Primer.tm), fn.Avg(
Primer.tm)).where(Primer.active == True).scalar(as_tuple=True))
nsets = Set.select(fn.Count(Set._id)).scalar()
if nsets > 0:
bs = Set.select().order_by(Set.score).limit(1).get()
bs_primers = ", ".join(bs.primer_seqs()).strip()
best_set = bs._id
bs_size = bs.set_size
bs_score = bs.score
bs_stats = "- " + "\n - ".join(
fmtkv(k, bs.__dict__['_data'][k])
for k in bs.exported_fields()
if k not in ["_id", "pids", "score", "primers"])
self.best_set_desc = self.best_set_desc.format(**locals())
if_no_primers_msg = click.style(
"Run `swga count` to find possible primers."
if nprimers == 0 else "",
fg='green')
if_no_active_primers_msg = click.style(
"Run `swga filter` to identify primers to use."
if nactive == 0 else "",
fg='green')
melting_tmp_msg = (
"The melting temp of the primers ranges between {min_tm:.2f}C and "
"{max_tm:.2f}C with an average of {avg_tm:.2f}C."
if nactive > 0 and min_tm and max_tm else
"No melting temps have been calculated yet.").format(**locals())
ifzero_sets_msg = click.style(
"Run `swga find_sets` after identifying valid primers to begin "
"collecting sets.\n",
fg='green')
set_msg = (self.best_set_desc if nsets > 0 else ifzero_sets_msg)
primer_db = os.path.abspath(self.primer_db)
nprimers = click.style(str(nprimers), bold=True, fg='blue')
nactive = click.style(str(nactive), bold=True, fg='blue')
nsets = click.style(str(nsets), bold=True, fg='blue')
self.header = click.style("swga v{}".format(__version__), fg='green')
# Copy all the relevant values into one dict
values = self.__dict__.copy()
values.update(locals())
# Format the summary message with all the calculated values
self.summary_msg = self.summary_msg.format(**values)
click.echo(quote(self.summary_msg, quote=" ", width=200))