def test_plotgenometracks():
roi = pkg_resources.resource_filename('janggu', 'resources/sample.bed')
bw_file = pkg_resources.resource_filename('janggu', 'resources/sample.bw')
cover = Cover.create_from_bigwig('coverage2',
bigwigfiles=bw_file,
roi=roi,
binsize=200,
stepsize=200,
resolution=50)
cover2 = Cover.create_from_bigwig('coverage2',
bigwigfiles=bw_file,
roi=roi,
binsize=200,
stepsize=200,
resolution=50)
a = plotGenomeTrack([cover,cover2],'chr1',16000,18000)
a = plotGenomeTrack(cover,'chr1',16000,18000)
def test_bigwig_store_whole_genome_option():
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bed_file = os.path.join(data_path, "sample.bed")
bwfile_ = os.path.join(data_path, "sample.bw")
cover1 = Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
regions=bed_file,
store_whole_genome=True,
binsize=200,
stepsize=200,
storage='ndarray')
cover2 = Cover.create_from_bigwig('test2',
bigwigfiles=bwfile_,
regions=bed_file,
store_whole_genome=False,
binsize=200,
stepsize=200,
storage='ndarray')
assert len(cover1) == 100
assert len(cover2) == len(cover1)
assert cover1.shape == (100, 200, 1, 1)
assert cover1.shape == cover2.shape
np.testing.assert_equal(cover1[:], cover2[:])
assert cover1[:].sum() == 1044.0
def test_channel_last_first():
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bed_file = os.path.join(data_path, "sample.bed")
bwfile_ = os.path.join(data_path, "sample.bw")
cover = Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
resolution=1,
binsize=200,
regions=bed_file,
store_whole_genome=True,
channel_last=True,
storage='ndarray')
assert cover.shape == (100, 200, 1, 1)
assert cover[0].shape == (1, 200, 1, 1)
cover1 = cover
cover = Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
resolution=1,
binsize=200,
regions=bed_file,
store_whole_genome=True,
channel_last=False,
storage='ndarray')
assert cover.shape == (100, 1, 200, 1)
assert cover[0].shape == (1, 1, 200, 1)
np.testing.assert_equal(cover1[0], np.transpose(cover[0], (0, 2, 3, 1)))
def test_load_cover_bigwig_resolutionNone(tmpdir):
os.environ['JANGGU_OUTPUT'] = tmpdir.strpath
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bwfile_ = os.path.join(data_path, "sample.bw")
bed_file = os.path.join(data_path, "sample.bed")
for store in ['ndarray', 'hdf5', 'sparse']:
# base pair binsize
print(store)
cover1 = Cover.create_from_bigwig(
"cov",
bigwigfiles=bwfile_,
roi=bed_file,
binsize=200, stepsize=200,
resolution=1,
storage=store, cache=True)
cover = Cover.create_from_bigwig(
"cov",
bigwigfiles=bwfile_,
roi=bed_file,
binsize=200, stepsize=200,
resolution=None,
storage=store, cache=True, datatags=['None'],
collapser='sum')
np.testing.assert_equal(len(cover), 100)
np.testing.assert_equal(cover.shape, (100, 1, 1, 1))
np.testing.assert_equal(cover1[:].sum(axis=1), cover[:].sum(axis=1))
def test_load_cover_bigwig_default(tmpdir):
os.environ['JANGGU_OUTPUT'] = tmpdir.strpath
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bwfile_ = os.path.join(data_path, "sample.bw")
gsfile_ = os.path.join(data_path, 'sample.chrom.sizes')
gsize = pandas.read_csv(gsfile_,
sep='\t',
names=['chr', 'length'],
index_col='chr').to_dict()['length']
bed_file = os.path.join(data_path, "sample.bed")
for store in ['ndarray', 'hdf5', 'sparse']:
# base pair binsize
print(store)
cover = Cover.create_from_bigwig("cov",
bigwigfiles=bwfile_,
regions=bed_file,
binsize=200,
stepsize=200,
genomesize=gsize,
storage=store,
cache=True)
np.testing.assert_equal(len(cover), 100)
np.testing.assert_equal(cover.shape, (100, 200, 1, 1))
# there is one read in the region
np.testing.assert_allclose(cover[4].sum(), 36.)
np.testing.assert_allclose(cover[52].sum(), 2 * 36.)
def get_data(params):
zscore = ZScore()
LABELS_TRAIN = ReduceDim(Cover.create_from_bam(
'geneexpr',
bamfiles=RNA.format(params['traincell'], params['trainrep']),
roi=ROI_INPUT_TRAIN,
flank=params['cageflank'],
conditions=['GeneExpr'],
resolution=None,
store_whole_genome=False,
storage='ndarray',
normalizer=[LogTransform(), zscore],
stranded=False,
cache=True),
aggregator="mean")
train_labels = LABELS_TRAIN
train_input = []
if params['inputs'] in ['dna_only', 'epi_dna']:
dnaflank = params['dnaflank']
order = params['order']
# DNA
DNA_TRAIN = Bioseq.create_from_refgenome('dna',
refgenome=REFGENOME,
roi=ROI_INPUT_TRAIN,
flank=dnaflank,
order=order,
cache=True,
store_whole_genome=False)
train_input += [DNA_TRAIN]
if params['inputs'] in ['epi_only', 'epi_dna']:
zscore = ZScore()
dnase_TRAIN = ReduceDim(Cover.create_from_bam(
'dnase',
bamfiles=DNASE.format(params['traincell']),
roi=ROI_INPUT_TRAIN,
flank=params['dnaseflank'],
resolution=None,
store_whole_genome=False,
normalizer=[LogTransform(), zscore],
cache=True),
aggregator="mean")
train_input += [dnase_TRAIN]
zscore = ZScore()
h3k4_TRAIN = ReduceDim(Cover.create_from_bigwig(
'h3k4',
bigwigfiles=[H3K4me3.format(params['traincell'])],
roi=ROI_INPUT_TRAIN,
flank=params['dnaseflank'],
store_whole_genome=False,
normalizer=[LogTransform(), zscore],
cache=True),
aggregator="mean")
train_input += [h3k4_TRAIN]
if len(train_input) == 0:
raise ValueError('no input')
return (train_input, train_labels)
def test_load_cover_bigwig_resolution1(tmpdir):
os.environ['JANGGU_OUTPUT'] = tmpdir.strpath
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bwfile_ = os.path.join(data_path, "sample.bw")
bed_file = os.path.join(data_path, "sample.bed")
for store in ['ndarray', 'hdf5', 'sparse']:
# base pair binsize
print(store)
cover = Cover.create_from_bigwig(
"cov",
bigwigfiles=bwfile_,
roi=bed_file,
binsize=200, stepsize=200,
resolution=1,
storage=store, cache=True)
np.testing.assert_equal(len(cover), 100)
np.testing.assert_equal(cover.shape, (100, 200, 1, 1))
# there is one read in the region 4
np.testing.assert_allclose(cover[4].sum(), 36)
np.testing.assert_equal(cover[4][0, :, 0, 0],
np.asarray(
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,
1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,
1., 1., 1., 1., 1., 1., 1., 1., 1., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]))
# and two reads in region 52
np.testing.assert_allclose(cover[52].sum(), 2*36)
np.testing.assert_equal(cover[52][0, :, 0, 0],
np.asarray(
[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 1., 1., 1., 1., 1., 1.,
1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,
1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,
1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,
1., 1., 1., 1., 1., 1., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,
0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]))
def test_bigwig_inferred_binsize():
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bed_file = os.path.join(data_path, "positive.bed")
bwfile_ = os.path.join(data_path, "sample.bw")
cover = Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
resolution=1,
regions=bed_file,
storage='ndarray')
assert len(cover) == 25
assert cover.shape == (25, 200, 1, 1)
def test_cover_export_bigwig(tmpdir):
path = tmpdir.strpath
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bed_file = os.path.join(data_path, "sample.bed")
bwfile_ = os.path.join(data_path, "sample.bw")
for resolution in [1, 50]:
for storage in [True, False]:
print('resolution=', resolution)
print('store_whole_genome', storage)
cover = Cover.create_from_bigwig(
'test',
bigwigfiles=bwfile_,
resolution=resolution,
binsize=200,
roi=bed_file,
store_whole_genome=storage,
storage='ndarray')
cover.export_to_bigwig(output_dir=path)
cov2 = Cover.create_from_bigwig('test',
bigwigfiles='{path}/{name}.{sample}.bigwig'.format(
path=path, name=cover.name,
sample=cover.conditions[0]),
resolution=resolution,
binsize=200,
roi=bed_file,
store_whole_genome=storage,
storage='ndarray')
assert cover.shape == (100, 200 // resolution, 1, 1)
assert cover.shape == cov2.shape
k = list(cover.garray.handle.keys())[0]
np.testing.assert_allclose(cover[:].sum(), 1044.0 / resolution)
np.testing.assert_allclose(cov2[:].sum(), 1044.0 / resolution)
def test_bigwig_genomic_interval_access_part_genome():
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bed_file = os.path.join(data_path, "sample.bed")
bamfile_ = os.path.join(data_path, "sample.bw")
storage = False
for reso in [1, 50]:
for shift in [0, 1]:
cover = Cover.create_from_bigwig(
'test',
bigwigfiles=bamfile_,
roi=bed_file,
flank=0,
storage='ndarray',
store_whole_genome=storage,
resolution=reso)
for i in range(len(cover)):
print('storage :',storage,'/ resolution :',reso,'/ shift :',shift)
print(i, cover.gindexer[i])
np.testing.assert_equal(np.repeat(cover[i],
cover.garray.resolution,
axis=1), cover[cover.gindexer[i]])
chrom, start, end, strand = cover.gindexer[i].chrom, \
cover.gindexer[i].start, \
cover.gindexer[i].end, \
cover.gindexer[i].strand
np.testing.assert_equal(np.repeat(cover[i],
cover.garray.resolution, axis=1),
cover[chrom, start, end, strand])
if shift != 0:
start += shift * reso
end += shift * reso
if strand != '-':
gicov = cover[chrom, start, end, strand][:, :(-shift*reso),:,:]
np.testing.assert_equal(cover[i][:, shift:,:, :],
gicov.reshape((1, gicov.shape[1]//reso, reso, 1, 1))[:, :, 0, :, :])
else:
gicov = cover[chrom, start, end, strand][:, (shift*reso):,:,:]
np.testing.assert_equal(cover[i][:, :-shift,:, :],
gicov.reshape((1, gicov.shape[1]//reso, reso, 1, 1))[:, :, 0, :, :])
def test_bigwig_genomic_interval_access():
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bed_file = os.path.join(data_path, "sample.bed")
bamfile_ = os.path.join(data_path, "sample.bw")
for storage in [True, False]:
for reso in [1, 50]:
for shift in [0, 1]:
cover = Cover.create_from_bigwig('test',
bigwigfiles=bamfile_,
regions=bed_file,
flank=0,
storage='ndarray',
store_whole_genome=storage,
resolution=reso)
for i in range(len(cover)):
print('storage :', storage, '/ resolution :', reso,
'/ shift :', shift)
print(i, cover.gindexer[i])
np.testing.assert_equal(cover[i], cover[cover.gindexer[i]])
chrom, start, end, strand = cover.gindexer[
i].chrom, cover.gindexer[i].start, cover.gindexer[
i].end, cover.gindexer[i].strand
np.testing.assert_equal(cover[i], cover[chrom, start, end,
strand])
if shift != 0:
start += shift * reso
end += shift * reso
if strand != '-':
np.testing.assert_equal(
cover[i][:, shift:, :, :],
cover[chrom, start, end,
strand][:, :-shift, :, :])
else:
np.testing.assert_equal(
cover[i][:, :-shift, :, :],
cover[chrom, start, end, strand][:,
shift:, :, :])
data, high, low = get_high_low_data(input, high_hbf, low_hbf)
roi_A, roi = get_roi(high + low)
seq = roi2fasta(roi_A, refgenome, flank)
test = pd.DataFrame.from_dict(seq, orient='index')
data['seq'] = test[0]
# 1. using janggu get DNA one-hot
## get one-hot data and ATAC feature matrix
dna_A = Bioseq.create_from_refgenome(name='dna',
refgenome=refgenome,
roi=roi_A,
flank=flank)
Tn5 = Cover.create_from_bigwig('bigwig_coverage',
bigwigfiles=bw_file,
roi=roi,
binsize=1,
stepsize=1,
flank=flank)
## ReShape
dna_A = np.reshape(dna_A, (len(high + low), flank * 2 + 1, 4))
bw_values = np.reshape(Tn5, (len(high + low), flank * 2 + 1))
## get motif PWM, 3. read meme get motif PWMs in both strands
motifs = read_motif(meme_file)
# 4. scan motifs get score_list, max(pos_strand,neg_strand)
score_list_A = Parallel(n_jobs=-1)(
delayed(DNA_motif_scan)(dna_A, motifs[m][0], motifs[m][1]) for m in motifs)
import matplotlib.pyplot as plt
from pkg_resources import resource_filename
from janggu.data import Cover
from janggu.data import plotGenomeTrack
roi = resource_filename('janggu', 'resources/sample.bed')
bw_file = resource_filename('janggu', 'resources/sample.bw')
cover = Cover.create_from_bigwig('coverage1',
bigwigfiles=[bw_file] * 2,
conditions=['rep1', 'rep2'],
roi=roi,
binsize=200,
stepsize=200,
resolution=50)
cover2 = Cover.create_from_bigwig('coverage2',
bigwigfiles=bw_file,
roi=roi,
binsize=200,
stepsize=200,
resolution=50)
a = plotGenomeTrack([cover, cover2], 'chr1', 16000, 18000)
a.savefig('coverage.png')
#plt.show(a)
def test_cover_from_bigwig_sanity():
data_path = pkg_resources.resource_filename('janggu', 'resources/')
bed_file = os.path.join(data_path, "sample.bed")
bwfile_ = os.path.join(data_path, "sample.bw")
cover = Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
regions=bed_file,
binsize=200,
stepsize=50,
resolution=50,
flank=0,
storage='ndarray')
cover[0]
assert len(cover.gindexer) == 394
assert len(cover.garray.handle) == 394
cover = Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
regions=bed_file,
binsize=200,
stepsize=50,
resolution=50,
flank=0,
storage='ndarray',
store_whole_genome=True)
cover[0]
assert len(cover.gindexer) == 394
assert len(cover.garray.handle) == 2
cov2 = Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
resolution=7,
storage='ndarray',
store_whole_genome=True)
assert len(cov2.garray.handle) == 2
assert cov2['chr1', 100, 200].shape == (1, 100 // 7 + 1, 1, 1)
with pytest.raises(Exception):
cov2.shape
with pytest.raises(Exception):
cov2[0]
with pytest.raises(Exception):
# name must be a string
Cover.create_from_bigwig(1.2,
bigwigfiles=bwfile_,
regions=bed_file,
binsize=1,
stepsize=1,
storage='ndarray')
with pytest.raises(Exception):
Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
regions=bed_file,
binsize=1,
stepsize=1,
flank=-1,
storage='ndarray')
with pytest.raises(Exception):
Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
regions=bed_file,
binsize=1,
stepsize=-1,
flank=0,
storage='ndarray')
with pytest.raises(Exception):
Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
regions=bed_file,
binsize=-1,
stepsize=1,
flank=0,
storage='ndarray')
with pytest.raises(Exception):
# resolution must be greater than stepsize
Cover.create_from_bigwig('test',
bigwigfiles=bwfile_,
regions=bed_file,
binsize=200,
stepsize=50,
resolution=300,
flank=0,
storage='ndarray')
