def slop_list2gff():
global_names=globals()
for protein_id in slop_list:
if protein_id[0:3]=="MGG":
head_id=protein_id
yield asinterval(MGG_db[protein_id[0:9]])
else:
strain_id,protein_ordinal=strain_protein_id_pattern.search(protein_id).group(1,2)
yield asinterval(global_names.get(strain_id+"_db")["gene_"+protein_ordinal])
def intron_generator():
"""
Construct intron features by subtracting all exons from all genes.
"""
genes = pybedtools.BedTool(
asinterval(g) for g in settings.G.features_of_type('gene')\
if g.chrom in settings.CHROMS)
exons = pybedtools.BedTool(
asinterval(e) for e in settings.G.features_of_type('exon')\
if e.chrom in settings.CHROMS)
for feature in genes.subtract(exons).saveas():
yield feature
def intron_generator():
"""
Construct intron features by subtracting all exons from all genes.
"""
genes = pybedtools.BedTool(
asinterval(g) for g in settings.G.features_of_type('gene')\
if g.chrom in settings.CHROMS)
exons = pybedtools.BedTool(
asinterval(e) for e in settings.G.features_of_type('exon')\
if e.chrom in settings.CHROMS)
for feature in genes.subtract(exons).saveas():
yield feature
def features(self, ignore_unknown=False):
"""
Generator of currently-selected features.
Looks up each feature in the attached `gffutils.FeatureDB` and converts
it into a `pybedtools.Interval` object for use with `pybedtools`.
Raises a warning if you haven't yet attached a `gffutils.FeatureDB` to
this instance.
:param ignore_unknown: If `ignore_unknown=False` then an exception will
be raised if a feature cannot be found; if `ignore_unknown=True`
then silently ignore these cases. Consider using the
`strip_unknown_features()` method to handle these cases up front.
"""
if not self.gffdb:
raise ValueError('Please attach a GFF database created by '
'gffutils by setting the .gffdb attribute to the '
'database\'s path.')
for i in self.data[self.id_column]:
try:
yield asinterval(self.gffdb[i])
except gffutils.FeatureNotFoundError:
if ignore_unknown:
continue
else:
raise gffutils.FeatureNotFoundError('%s not found' % i.id)
def features(self, ignore_unknown=False):
"""
Generator of currently-selected features.
Looks up each feature in the attached `gffutils.FeatureDB` and converts
it into a `pybedtools.Interval` object for use with `pybedtools`.
Raises a warning if you haven't yet attached a `gffutils.FeatureDB` to
this instance.
:param ignore_unknown: If `ignore_unknown=False` then an exception will
be raised if a feature cannot be found; if `ignore_unknown=True`
then silently ignore these cases. Consider using the
`strip_unknown_features()` method to handle these cases up front.
"""
if not self.gffdb:
raise ValueError('Please attach a GFF database created by '
'gffutils by setting the .gffdb attribute to the '
'database\'s path.')
for i in self.data[self.id_column]:
try:
yield asinterval(self.gffdb[i])
except gffutils.FeatureNotFoundError:
if ignore_unknown:
continue
else:
raise gffutils.FeatureNotFoundError('%s not found' % i.id)
def _make_track(self, d, cls):
yheight = self.heights[cls]
ybase = self.ybase + (self.heights['full'] - yheight) * 0.5
return Track((asinterval(i) for i in d[cls]),
ybase=ybase,
yheight=yheight,
**self.kwargs)
def gen():
"""
Generator of pybedtools.Intervals representing TSSes.
"""
for gene in db.features_of_type('gene'):
for transcript in db.children(gene, level=1):
if transcript.strand == '-':
transcript.start = transcript.stop
else:
transcript.stop = transcript.start
transcript.featuretype = transcript.featuretype + '_TSS'
yield helpers.asinterval(transcript)
def gen():
"""
Generator of pybedtools.Intervals representing TSSes.
"""
for gene in db.features_of_type('gene'):
for transcript in db.children(gene, level=1):
if transcript.strand == '-':
transcript.start = transcript.stop
else:
transcript.stop = transcript.start
transcript.featuretype = transcript.featuretype + '_TSS'
yield helpers.asinterval(transcript)
def gene_generator():
"""
The database has inferred full gene models from the GTF, so we can simply
iterate over them here.
More complex generators can be created as well -- for example, one that
only returns unique TSS sites from all isoforms of all genes.
"""
for g in settings.G.features_of_type('gene'):
if g.chrom not in settings.CHROMS:
continue
yield asinterval(g)
def gene_generator():
"""
The database has inferred full gene models from the GTF, so we can simply
iterate over them here.
More complex generators can be created as well -- for example, one that
only returns unique TSS sites from all isoforms of all genes.
"""
for g in settings.G.features_of_type('gene'):
if g.chrom not in settings.CHROMS:
continue
yield asinterval(g)
def test_pbt_interval_conversion():
try:
import pybedtools
except ImportError:
return
line = "chr2L FlyBase exon 7529 8116 . + . Name=CG11023:1;Parent=FBtr0300689,FBtr0300690"
f = feature.feature_from_line(line, strict=False, keep_order=True)
pbt = helpers.asinterval(f)
assert pbt.chrom == f.chrom == f.seqid
assert pbt.start == f.start - 1
assert pbt.stop == f.stop == f.end
pn = pbt.name
fn = f.attributes['Name'][0]
assert pn == fn, '%s, %s' % (pn, fn)
def test_pbt_interval_conversion():
try:
import pybedtools
except ImportError:
return
line = "chr2L FlyBase exon 7529 8116 . + . Name=CG11023:1;Parent=FBtr0300689,FBtr0300690"
f = feature.feature_from_line(line, strict=False, keep_order=True)
pbt = helpers.asinterval(f)
assert pbt.chrom == f.chrom == f.seqid
assert pbt.start == f.start -1
assert pbt.stop == f.stop == f.end
pn = pbt.name
fn = f.attributes['Name'][0]
assert pn == fn, '%s, %s' % (pn, fn)
def plot(self, feature):
"""
Spawns a new figure showing data for `feature`.
:param feature: A `pybedtools.Interval` object
Using the pybedtools.Interval `feature`, creates figure specified in
:meth:`BaseMiniBrowser.make_fig` and plots data on panels according to
`self.panels()`.
"""
if isinstance(feature, gffutils.Feature):
feature = asinterval(feature)
self.make_fig()
axes = []
for ax, method in self.panels():
feature = method(ax, feature)
axes.append(ax)
return axes
def plot(self, feature):
"""
Spawns a new figure showing data for `feature`.
:param feature: A `pybedtools.Interval` object
Using the pybedtools.Interval `feature`, creates figure specified in
:meth:`BaseMiniBrowser.make_fig` and plots data on panels according to
`self.panels()`.
"""
if isinstance(feature, gffutils.Feature):
feature = asinterval(feature)
self.make_fig()
axes = []
for ax, method in self.panels():
feature = method(ax, feature)
axes.append(ax)
return axes
def TSS(feature, upstream=1000, downstream=1000):
"""
Transforms a pybedtools.Interval, `feature`, into a TSS extended by
upstream/downstream, paying attention to strand and proximity to chromosome
limits.
Also edits the feature type to be "TSS"
"""
chrom_size = chromsizes[feature.chrom][1]
if feature.strand == '-':
start = max(0, feature.stop - downstream)
stop = min(feature.stop + upstream, chrom_size)
else:
start = max(0, feature.start - upstream)
stop = min(feature.start + downstream, chrom_size)
# Modify featuretype
feature[2] = 'TSS'
feature.start = start
feature.stop = stop
return asinterval(feature)
def TSS(feature, upstream=1000, downstream=1000):
"""
Transforms a pybedtools.Interval, `feature`, into a TSS extended by
upstream/downstream, paying attention to strand and proximity to chromosome
limits.
Also edits the feature type to be "TSS"
"""
chrom_size = chromsizes[feature.chrom][1]
if feature.strand == '-':
start = max(0, feature.stop - downstream)
stop = min(feature.stop + upstream, chrom_size)
else:
start = max(0, feature.start - upstream)
stop = min(feature.start + downstream, chrom_size)
# Modify featuretype
feature[2] = 'TSS'
feature.start = start
feature.stop = stop
return asinterval(feature)
def scored_feature_generator(d):
for i in range(len(d)):
try:
feature = db[d.id[i]]
except gffutils.FeatureNotFoundError:
raise gffutils.FeatureNotFoundError(d.id[i])
score = -10 * np.log10(d.padj[i])
lfc = d.log2foldchange[i]
if np.isnan(lfc):
score = 0
if lfc < 0:
score *= -1
feature.score = str(score)
feature = extend_fields(gff2bed(asinterval(feature)), 9)
fields = feature.fields[:]
fields[6] = fields[1]
fields[7] = fields[2]
fields.append(str(d.padj[i]))
fields.append(str(d.pval[i]))
fields.append('%.3f' % d.log2foldchange[i])
fields.append('%.3f' % d.basemeana[i])
fields.append('%.3f' % d.basemeanb[i])
yield pybedtools.create_interval_from_list(fields)
def scored_feature_generator(d):
for i in range(len(d)):
try:
feature = db[d.id[i]]
except gffutils.FeatureNotFoundError:
raise gffutils.FeatureNotFoundError(d.id[i])
score = -10 * np.log10(d.padj[i])
lfc = d.log2foldchange[i]
if np.isnan(lfc):
score = 0
if lfc < 0:
score *= -1
feature.score = str(score)
feature = extend_fields(gff2bed(asinterval(feature)), 9)
fields = feature.fields[:]
fields[6] = fields[1]
fields[7] = fields[2]
fields.append(str(d.padj[i]))
fields.append(str(d.pval[i]))
fields.append('%.3f' % d.log2foldchange[i])
fields.append('%.3f' % d.basemeana[i])
fields.append('%.3f' % d.basemeanb[i])
yield pybedtools.create_interval_from_list(fields)
def tss_generator():
for transcript in db.features_of_type("transcript"):
yield TSS(asinterval(transcript), upstream=1000, downstream=1000)
def generator():
for gene_id in df.index:
yield asinterval(db[gene_id])
try: # if this transcript has an entry for 'transcript_support_level'
# and if the level is below the acceptable threshold:
# keep that transcript
if int([
i[1] for i in t.attributes.items()
if i[0] == 'transcript_support_level'
][0][0]) <= max_TSL:
temp_txpts.append(t)
except:
pass
txpts = temp_txpts
if (len(txpts) > 0
): # if there are any transcripts with a sufficiently low TSL:
all_exons = (pybedtools.BedTool([
helpers.asinterval(i)
for i in db.children(gene, featuretype='exon')
]))
all_exons = all_exons.sort().merge(
) # define the ends of the genic region by using the first
# and last exon in the annotation as the limits
gene_extent = pybedtools.BedTool([
pybedtools.cbedtools.Interval(chrom=chrom,
start=min(i.start
for i in all_exons),
end=max(i.end
for i in all_exons))
])
t_introns = []
for t in txpts: # for each transcript, get all of the exons
t_exons = (pybedtools.BedTool([
def generate_interval(category_fl):
for strain_id_raw in category_fl:
strain_id = strain_id_raw.strip('\n')
if strain_id[0:3] != "MGG": continue
yield gff2bed(asinterval(MGG_db[strain_id]), name_field=2)
def generator():
G = gffutils.FeatureDB(dbfn)
genes = G.features_of_type('gene')
for i in range(5000):
yield asinterval(genes.next())
def _make_track(self, d, cls):
yheight = self.heights[cls]
ybase = self.ybase + (self.heights['full'] - yheight) * 0.5
return Track(
(asinterval(i) for i in d[cls]),
ybase=ybase, yheight=yheight, **self.kwargs)
def tss_generator():
for transcript in db.features_of_type('mRNA'): #CDS/gene/mRNA...
yield TSS(asinterval(transcript), upstream=1, downstream=0)
def generator():
G = gffutils.FeatureDB(dbfn)
genes = G.features_of_type('gene')
for i in range(5000):
yield asinterval(genes.next())
def tss_generator():
"""
Generator function to yield TSS of each annotated transcript
"""
for transcript in db.features_of_type('transcript'):
yield TSS(asinterval(transcript), upstream=1, downstream=0)
def generate_bed(gff_feature_item):
yield asinterval(gff_feature_item)
def gen():
for i in iterator:
yield helpers.asinterval(i)
def generator():
for gene_id in df.index:
yield asinterval(db[gene_id])
def tss_generator():
for transcript in db.features_of_type('transcript'):
yield TSS(asinterval(transcript), upstream=1000, downstream=1000)
def fsdu(which_id):
yield asinterval(db[which_id])
def gen():
for i in iterator:
yield helpers.asinterval(i)
def tss_generator(gtf):
"""
Generator function to yield TSS +/- 1kb of each annotated transcript
"""
for transcript in db.features_of_type('transcript'):
yield TSS(asinterval(transcript), upstream=1000, downstream=1000)
