def runTest(self):
for num, info in sorted(samples['small_signals'].items()):
# Prepare paths #
orig_bigwig_path = info['bigwig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_bigwig_path = temporary_path('.bigwig')
# From bigwig to SQL #
try:
track.convert(orig_bigwig_path, test_sql_path, assembly='sacCer2')
except MissingExecutableError as err:
sys.stderr.write("skipping: {0} ".format(err.message))
break
else:
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to bigwig #
try:
track.convert(test_sql_path, test_bigwig_path)
except MissingExecutableError as err:
sys.stderr.write("skipping: {0} ".format(err.message))
break
else:
self.assertTrue(assert_file_equal(orig_bigwig_path, test_bigwig_path, start_a=1, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_bigwig_path)
def runTest(self):
for num, info in sorted(samples['small_signals'].items()):
# Prepare paths #
orig_bigwig_path = info['bigwig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_bigwig_path = temporary_path('.bigwig')
# From bigwig to SQL #
try:
track.convert(orig_bigwig_path,
test_sql_path,
assembly='sacCer2')
except MissingExecutableError as err:
sys.stderr.write("skipping: {0} ".format(err.message))
break
else:
self.assertTrue(assert_file_equal(orig_sql_path,
test_sql_path))
# From SQL to bigwig #
try:
track.convert(test_sql_path, test_bigwig_path)
except MissingExecutableError as err:
sys.stderr.write("skipping: {0} ".format(err.message))
break
else:
self.assertTrue(
assert_file_equal(orig_bigwig_path,
test_bigwig_path,
start_a=1,
start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_bigwig_path)
def serialize_chr_file(chrmeta, path=None, seperator='\t'):
"""Read a dictionary and create a plain text file"""
if not path: path = temporary_path()
def lines():
for k,v in chrmeta.items(): yield k + seperator + str(v['length']) + '\n'
with open(path, 'w') as f: f.writelines(lines())
return path
def runTest(self):
out_path = temporary_path('.sql')
with track.new(out_path) as t:
for chrom in range(5): t.write(str(chrom), [(0,10,'A',0.0,-1)])
cur = t.cursor()
cur.execute("CREATE table tmp (koopa text,troopa text)")
cur.execute("INSERT into tmp values (?,?)", (1,2))
def runTest(self):
for num, info in sorted(samples['small_signals'].items()):
# Prepare paths #
orig_bigwig_path = info['bigwig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_bigwig_path = temporary_path('.bigwig')
# From bigwig to SQL #
track.convert(orig_bigwig_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to bigwig #
track.convert(test_sql_path, test_bigwig_path)
self.assertTrue(assert_file_equal(orig_bigwig_path, test_bigwig_path, start_a=1, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_bigwig_path)
def runTest(self):
out_path = temporary_path('.sql')
with track.new(out_path) as t:
for chrom in range(5):
t.write(str(chrom), [(0, 10, 'A', 0.0, -1)])
cur = t.cursor()
cur.execute("CREATE table tmp (koopa text,troopa text)")
cur.execute("INSERT into tmp values (?,?)", (1, 2))
def runTest(self):
for num, info in sorted(samples['small_features'].items()):
# Prepare paths #
orig_bed_path = info['bed']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_bed_path = temporary_path('.bed')
# From BED to SQL #
track.convert(orig_bed_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to BED #
with track.load(test_sql_path) as t: t.roman_to_integer()
track.convert(test_sql_path, test_bed_path)
self.assertTrue(assert_file_equal(orig_bed_path, test_bed_path, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_bed_path)
def runTest(self):
for num, info in sorted(samples['gtf_tracks'].items()):
if num == 'GenRep': continue
orig_gtf_path = info['gtf']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_gtf_path = temporary_path('.gtf')
# From GTF to SQL #
track.convert(orig_gtf_path, test_sql_path)
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to GTF #
# Use Track() instead. #
track.convert(test_sql_path, test_gtf_path)
self.assertTrue(assert_file_equal(orig_gtf_path, test_gtf_path, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_gtf_path)
def runTest(self):
in_path = samples['small_signals'][7]['sql']
out_path = temporary_path('.sql')
with track.load(in_path) as i:
with track.new(out_path) as o:
for chrom in i: o.write(chrom, i.read(chrom))
self.assertEqual(list(o.read('chrI')), list(i.read('chrI')))
os.remove(out_path)
def runTest(self):
for num, info in sorted(samples['gtf_tracks'].items()):
if num == 'GenRep': continue
orig_gtf_path = info['gtf']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_gtf_path = temporary_path('.gtf')
# From GTF to SQL #
track.convert(orig_gtf_path, test_sql_path)
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to GTF #
# Use Track() instead. #
track.convert(test_sql_path, test_gtf_path)
self.assertTrue(
assert_file_equal(orig_gtf_path, test_gtf_path, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_gtf_path)
def runTest(self):
in_path = samples['small_signals'][7]['sql']
out_path = temporary_path('.sql')
with track.load(in_path) as i:
with track.new(out_path) as o:
for chrom in i:
o.write(chrom, i.read(chrom))
self.assertEqual(list(o.read('chrI')), list(i.read('chrI')))
os.remove(out_path)
def runTest(self):
for num, info in sorted(samples["gff_tracks"].items()):
# Prepare paths #
orig_gff_path = info["gff"]
orig_sql_path = info["sql"]
test_sql_path = temporary_path(".sql")
test_gff_path = temporary_path(".gff")
# From GFF to SQL #
track.convert(orig_gff_path, test_sql_path, assembly="sacCer2")
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to GFF #
with track.load(test_sql_path) as t:
t.roman_to_integer()
track.convert(test_sql_path, test_gff_path)
self.assertTrue(assert_file_equal(orig_gff_path, test_gff_path, start_a=1, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_gff_path)
def runTest(self):
for num, info in sorted(samples['small_signals'].items()):
# Some files cannot be roundtriped #
if num == 3 or num == 7: continue
# Prepare paths #
orig_wig_path = info['wig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_wig_path = temporary_path('.wig')
# From WIG to SQL #
track.convert(orig_wig_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to WIG #
with track.load(test_sql_path) as t: t.roman_to_integer()
track.convert(test_sql_path, test_wig_path)
self.assertTrue(assert_file_equal(orig_wig_path, test_wig_path, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_wig_path)
def runTest(self):
for num, info in sorted(samples['small_features'].items()):
# Prepare paths #
orig_bed_path = info['bed']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_bed_path = temporary_path('.bed')
# From BED to SQL #
track.convert(orig_bed_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to BED #
with track.load(test_sql_path) as t:
t.roman_to_integer()
track.convert(test_sql_path, test_bed_path)
self.assertTrue(
assert_file_equal(orig_bed_path, test_bed_path, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_bed_path)
def runTest(self):
# Prepare paths #
orig_gzip_path = samples['gzip_tracks'][1]['gzip']
orig_sql_path = samples['small_features'][1]['sql']
test_sql_path = temporary_path('.sql')
# From BED to SQL #
track.convert(orig_gzip_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def runTest(self):
for num, info in sorted(samples['rand_signals'].items()):
# Prepare paths #
orig_wig_path = info['wig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
# From WIG to SQL #
track.convert(orig_wig_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def runTest(self):
for num, info in sorted(samples['yeast_features'].items()):
# Prepare paths #
orig_bed_path = info['bed']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
# From BED to SQL #
track.convert(orig_bed_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def runTest(self):
in_path = samples['small_signals'][4]['sql']
out_path = temporary_path('.sql')
t = threshold(in_path, 8000.0)
t.export(out_path)
with track.load(out_path) as t:
data = t.read('chrI')
got = map(tuple, data)
expected = [(120, 122, 9000.0)]
self.assertEqual(got, expected)
os.remove(out_path)
def runTest(self):
for num, info in sorted(samples['yeast_features'].items()):
# Prepare paths #
orig_bed_path = info['bed']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
# From BED to SQL #
track.convert(orig_bed_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def runTest(self):
for num, info in sorted(samples['sga_tracks'].items()):
# Prepare paths #
orig_sga_path = info['sga']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
# From SGA to SQL #
track.convert(orig_sga_path, test_sql_path, assembly='hg19')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def runTest(self):
for num, info in sorted(samples['small_signals'].items()):
# Prepare paths #
orig_bigwig_path = info['bigwig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_bigwig_path = temporary_path('.bigwig')
# From bigwig to SQL #
track.convert(orig_bigwig_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to bigwig #
track.convert(test_sql_path, test_bigwig_path)
self.assertTrue(
assert_file_equal(orig_bigwig_path,
test_bigwig_path,
start_a=1,
start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_bigwig_path)
def runTest(self):
info = samples['small_signals'][1]
# Prepare paths #
orig_sql_path = info['sql']
orig_sga_path = info['sga']
test_sga_path = temporary_path('.sga')
# From SGA to SQL #
track.convert(orig_sql_path, test_sga_path, assembly='hg19')
self.assertTrue(assert_file_equal(orig_sga_path, test_sga_path))
# Clean up #
os.remove(test_sga_path)
def serialize_chr_file(chrmeta, path=None, seperator='\t'):
"""Read a dictionary and create a plain text file"""
if not path: path = temporary_path()
def lines():
for k, v in chrmeta.items():
yield k + seperator + str(v['length']) + '\n'
with open(path, 'w') as f:
f.writelines(lines())
return path
def runTest(self):
for num, info in sorted(samples['rand_signals'].items()):
# Prepare paths #
orig_wig_path = info['wig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
# From WIG to SQL #
track.convert(orig_wig_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def runTest(self):
in_path = samples['small_features'][1]['sql']
out_path = temporary_path('.sql')
chrom = 'chrI'
with track.load(in_path) as i:
with track.new(out_path) as o:
o.fields = track.default_fields
o.write(chrom, i.read(chrom, ('start','end')))
got = tuple(o.read(chrom).next())
expected = (0, 10, None, None, None)
self.assertEqual(got, expected)
os.remove(out_path)
def runTest(self):
in_path = temporary_path('.sql')
out_path = temporary_path('.sql')
with track.new(in_path) as t:
t.fields = ('start','end','score')
t.assembly = 'sacCer2'
t.write('chrI',[(0,2,10),(2,4,20),(6,8,10)])
result = window_smoothing(t, 2)
result.export(out_path)
with track.load(out_path) as t:
data = t.read('chrI')
got = map(tuple, data)
expected = [(0, 1, 8.0),
(1, 3, 12.0),
(3, 5, 10.0),
(5, 6, 8.0),
(6, 9, 4.0),
(9, 10, 2.0)]
self.assertEqual(got, expected)
os.remove(in_path)
os.remove(out_path)
def runTest(self):
for num, info in sorted(samples['small_signals'].items()):
# Some files cannot be roundtriped #
if num == 3 or num == 7: continue
# Prepare paths #
orig_wig_path = info['wig']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_wig_path = temporary_path('.wig')
# From WIG to SQL #
track.convert(orig_wig_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to WIG #
with track.load(test_sql_path) as t:
t.roman_to_integer()
track.convert(test_sql_path, test_wig_path)
self.assertTrue(
assert_file_equal(orig_wig_path, test_wig_path, start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_wig_path)
def runTest(self):
in_path = samples['small_features'][1]['sql']
out_path = temporary_path('.sql')
chrom = 'chrI'
with track.load(in_path) as i:
with track.new(out_path) as o:
o.fields = track.default_fields
o.write(chrom, i.read(chrom, ('start', 'end')))
got = tuple(o.read(chrom).next())
expected = (0, 10, None, None, None)
self.assertEqual(got, expected)
os.remove(out_path)
def runTest(self):
x_path = samples['small_signals'][4]['sql']
y_path = samples['small_features'][4]['sql']
out_path = temporary_path('.sql')
t = mean_score_by_feature(x_path,y_path)
t.export(out_path)
with track.load(out_path) as t:
data = t.read('chrI')
got = map(tuple, data)
expected = [(10, 20, 15.0, u'Lorem', 1),
(30, 40, 50.0, u'Ipsum', 1)]
self.assertEqual(got, expected)
os.remove(out_path)
def runTest(self):
for num, info in sorted(samples['gtf_tracks'].items()):
# This one is too large #
if num == 'GenRep': continue
# Prepare paths #
orig_gtf_path = info['gtf']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
# From GTF to SQL #
track.convert(orig_gtf_path, test_sql_path)
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def runTest(self):
for num, info in sorted(samples['gff_tracks'].items()):
# Prepare paths #
orig_gff_path = info['gff']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
test_gff_path = temporary_path('.gff')
# From GFF to SQL #
track.convert(orig_gff_path, test_sql_path, assembly='sacCer2')
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# From SQL to GFF #
with track.load(test_sql_path) as t:
t.roman_to_integer()
track.convert(test_sql_path, test_gff_path)
self.assertTrue(
assert_file_equal(orig_gff_path,
test_gff_path,
start_a=1,
start_b=1))
# Clean up #
os.remove(test_sql_path)
os.remove(test_gff_path)
def runTest(self):
for num, info in sorted(samples['gtf_tracks'].items()):
# This one is too large #
if num == 'GenRep': continue
# Prepare paths #
orig_gtf_path = info['gtf']
orig_sql_path = info['sql']
test_sql_path = temporary_path('.sql')
# From GTF to SQL #
track.convert(orig_gtf_path, test_sql_path)
self.assertTrue(assert_file_equal(orig_sql_path, test_sql_path))
# Clean up #
os.remove(test_sql_path)
def parse(self):
# Special case #
if self.handler.format == 'bedGraph': return
# The two paths #
bigwig_path = self.path
bedgraph_path = temporary_path('.bedgraph')
# Run the tool #
bigwig_to_bedgraph(bigwig_path, bedgraph_path)
# Now just parse the bedgraph #
parser = track.parse.get_parser(bedgraph_path, 'bedgraph')
parser.name = self.name
parser(self.handler)
# Erase the temporary file #
os.remove(bedgraph_path)
def runTest(self):
orig_path = samples['small_features'][2]['sql']
test_path = temporary_path('.sql')
shutil.copy(orig_path, test_path)
with track.load(test_path) as t:
t.delete_fields(['name', 'strand'])
got = list(t.read())
expected = [('chrI', 10, 20, 0.1), ('chrI', 30, 40, 0.2),
('chrI', 50, 60, 0.1), ('chrI', 70, 80, 0.2),
('chrI', 90, 100, 0.0), ('chrI', 110, 120, 0.4),
('chrI', 130, 150, 0.4), ('chrI', 180, 190, 0.1),
('chrI', 180, 200, 0.1), ('chrI', 210, 220, 0.2),
('chrI', 230, 240, 0.1), ('chrI', 250, 260, 0.2),
('chrI', 270, 280, 0.0), ('chrI', 290, 300, 0.7)]
self.assertEqual(got, expected)
# Clean up #
os.remove(test_path)
def runTest(self):
in_path = samples['small_features'][1]['sql']
out_path = temporary_path('.bed')
with track.load(in_path) as t:
result = complement(t)
result.export(out_path)
with track.load(out_path) as t:
data = t.read('chrI')
got = map(tuple, data)
expected = [( 10, 20),
( 30, 40),
( 50, 60),
( 80, 90),
(110, 120),
(135, 230208)]
self.assertEqual(got, expected)
os.remove(out_path)
def bedgraph_to_bigwig(bedgraph_path, chrmeta, bigwig_path):
"""
Converts a bedgraph file to a bigwig file.
:param bedgraph_path: The path to the bedgraph file to read.
:type bedgraph_path: string
:param chrmeta: The chromosome metadata.
:type chrmeta: dict
:param bigwig_path: The path to the bedbig file to create.
:type bigwig_path: string
:returns: None
"""
# Check the binary tool exists #
check_executable('bedGraphToBigWig')
# Make the chr file #
chrfile_path = temporary_path('.chr')
serialize_chr_file(chrmeta, chrfile_path)
# Run the tool #
run_tool('bedGraphToBigWig', [bedgraph_path, chrfile_path, bigwig_path])
# Remove the chr file #
os.remove(chrfile_path)
def runTest(self):
in_paths = [samples['small_signals'][1]['sql'],
samples['small_signals'][2]['sql'],
samples['small_signals'][3]['sql']]
out_path = temporary_path('.sql')
t = merge_scores(in_paths)
t.export(out_path)
with track.load(out_path) as t:
data = t.read('chrI')
got = map(tuple, data)
expected = [( 0, 5, 2.0 + 0.6666666666666666),
( 5, 10, 4.0),
( 20, 30, 10.0),
( 30, 40, 30.0),
( 40, 50, 26.0 + 0.666666666666666),
( 50, 60, 120.0),
( 60, 68, 100.0),
( 68, 70, 200.0),
( 70, 80, 100.0),
( 90, 110, 3.0),
(120, 130, 10.0)]
self.assertEqual(got, expected)
os.remove(out_path)
def runTest(self):
orig_path = samples['small_features'][2]['sql']
test_path = temporary_path('.sql')
shutil.copy(orig_path, test_path)
with track.load(test_path) as t:
t.delete_fields(['name','strand'])
got = list(t.read())
expected = [('chrI', 10, 20, 0.1),
('chrI', 30, 40, 0.2),
('chrI', 50, 60, 0.1),
('chrI', 70, 80, 0.2),
('chrI', 90, 100, 0.0),
('chrI', 110, 120, 0.4),
('chrI', 130, 150, 0.4),
('chrI', 180, 190, 0.1),
('chrI', 180, 200, 0.1),
('chrI', 210, 220, 0.2),
('chrI', 230, 240, 0.1),
('chrI', 250, 260, 0.2),
('chrI', 270, 280, 0.0),
('chrI', 290, 300, 0.7)]
self.assertEqual(got, expected)
# Clean up #
os.remove(test_path)
def __enter__(self):
# Just serialize it as a bedgraph first #
self.tmp_path = temporary_path('.bedgraph')
self.file = open(self.tmp_path, 'w')
return self
def runTest(self):
in_path = samples['small_features'][1]['sql']
out_path = temporary_path('.bed')
track.convert(in_path, out_path)
#assert_file_equal(out_path, samples['small_features'][1]['bed'])
os.remove(out_path)
def runTest(self):
in_path = samples['small_features'][1]['sql']
out_path = temporary_path('.bed')
track.convert(in_path, out_path)
#assert_file_equal(out_path, samples['small_features'][1]['bed'])
os.remove(out_path)
