def test_split_by_strands():
genset = GenCoorSet(name="Test_set")
bedfile = os.path.join(os.getenv("HOME"),
"gencoor_data/hg38/genes_hg38.bed")
genset.load(filename=bedfile, filetype="BED")
res = genset.split_by_strands()
assert set([g.strand for g in res["+"]]) == set(["+"])
assert set([g.strand for g in res["-"]]) == set(["-"])
def test_standard_chromosome1():
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=50, name="test",
strand="."))
genset1.add(
GenCoor(chrom="chr2", start=20, end=40, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr3_random", start=70, end=80, name="test",
strand="."))
genset1.standard_chromosome()
assert len(genset1) == 2
def test_rm_duplicates2():
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset1.add(GenCoor(chrom="chr2", start=2, end=4, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr3_random", start=1, end=80, name="test", strand="."))
genset1.add(GenCoor(chrom="chr2", start=2, end=4, name="test2",
strand="-"))
res = genset1.rm_duplicates(inplace=False)
assert len(res) == 4
def test_GenCoorSet_len():
genset = GenCoorSet(name="Test_set")
genset.add(GenCoor(chrom="chr1", start=10, end=20, name="test",
strand="."))
genset.add(GenCoor(chrom="chr1", start=15, end=50, name="test",
strand="."))
genset.add(
GenCoor(chrom="chr2", start=100, end=200, name="test", strand="."))
assert len(genset) == 3
def test_merge():
genset = GenCoorSet(name="Test_set")
genset.add(GenCoor(chrom="chr1", start=10, end=20, name="test",
strand="+"))
genset.add(GenCoor(chrom="chr1", start=15, end=50, name="test",
strand="-"))
genset.add(
GenCoor(chrom="chr2", start=100, end=200, name="test", strand="."))
res = genset.merge(w_return=True)
assert len(res) == 2
assert res[0].start == 10
assert res[0].end == 50
assert res[0].strand == "."
res = genset.merge(w_return=True, strand_specific=True)
assert len(res) == 3
assert res[0].start == 10
assert res[0].end == 20
assert res[0].strand == "+"
def test_total_coverage1():
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset1.add(GenCoor(chrom="chr2", start=2, end=4, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr3_random", start=1, end=80, name="test", strand="."))
cov = genset1.total_coverage()
assert cov == 85
def test_distance2():
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=20, end=24, name="test", strand="."))
g = GenCoor(chrom="chr1", start=7, end=19, name="test", strand=".")
res = genset1.distance(g, sign=False)
assert res == 1
def test_intersect_3():
"""
A : none
B : -----
R : none
"""
genset1 = GenCoorSet(name="Test_set")
genset2 = GenCoorSet(name="Test_set")
genset2.add(
GenCoor(chrom="chr1", start=10, end=20, name="test", strand="+"))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 0
res = genset1.intersect(genset2, mode="original")
assert len(res) == 0
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def test_GenCoorSet_load():
genset = GenCoorSet(name="Test_set")
bedfile = os.path.join(os.getenv("HOME"),
"gencoor_data/hg38/genes_hg38.bed")
bedfile2 = os.path.join(os.getenv("HOME"),
"gencoor_data/hg38/genes_hg38_test.bed")
genset.load(filename=bedfile, filetype="BED")
genset.save(filename=bedfile2, filetype="BED")
os.remove(bedfile2)
def test_intersect_1():
"""
Two empty sets
A : none
B : none
R : none
"""
genset1 = GenCoorSet(name="Test_set")
genset2 = GenCoorSet(name="Test_set")
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 0
res = genset1.intersect(genset2, mode="original")
assert len(res) == 0
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def bed_sig2arr(bedname, sig_names, exp, arg):
regions = GenCoorSet(name=bedname)
regions.load(filename=exp.get_file(bedname))
if arg["--test"]:
regions.list = regions.list[0:50]
regions.relocate(mode='center as center', width=2*int(arg["--ext"]))
sig = SignalProfile(regions, genome=arg["--genome"],
bin=int(arg["--bin"]), step=int(arg["--step"]),
cores=int(arg["--cores"]))
for signal in sig_names:
if exp.get_file(signal).endswith(".bw") or exp.get_file(signal).endswith(".bigwig"):
sig.load_bigwig(filename=exp.get_file(signal), label=signal,
disable_progressbar=False, verbal=False)
elif exp.get_file(signal).endswith(".bam"):
sig.load_bam(filename=exp.get_file(signal), label=signal,
disable_progressbar=False, verbal=False)
res = sig.cov2array()
return res
def test_intersect_7():
"""
Perfect overlapping
A : ------
B : ------
R : ------
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=10, name="test",
strand="."))
genset1.add(
GenCoor(chrom="chr1", start=500, end=550, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=600, end=650, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=700, end=750, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=725, end=800, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=1, end=10, name="test",
strand="."))
genset2.add(
GenCoor(chrom="chr1", start=500, end=550, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=600, end=650, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=700, end=750, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=725, end=800, name="test", strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 6
res = genset1.intersect(genset2, mode="original")
assert len(res) == 5
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 5
def test_total_coverage2():
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
cov = genset1.total_coverage()
assert cov == 4
def test_extend():
genset = GenCoorSet(name="Test_set")
genset.add(GenCoor(chrom="chr1", start=10, end=20, name="test",
strand="+"))
genset.add(GenCoor(chrom="chr1", start=15, end=50, name="test",
strand="-"))
genset.add(
GenCoor(chrom="chr2", start=100, end=200, name="test", strand="."))
ngcs = genset.extend(mode="left", length=5)
assert ngcs.list[0].start == 5
assert ngcs.list[0].end == 20
assert ngcs.list[1].start == 10
assert ngcs.list[1].end == 50
ngcs = genset.extend(mode="right", length=5)
assert ngcs.list[0].start == 10
assert ngcs.list[0].end == 25
assert ngcs.list[1].start == 15
assert ngcs.list[1].end == 55
ngcs = genset.extend(mode="5end", length=5)
assert ngcs.list[0].start == 5
assert ngcs.list[0].end == 20
assert ngcs.list[1].start == 15
assert ngcs.list[1].end == 55
ngcs = genset.extend(mode="3end", length=5)
assert ngcs.list[0].start == 10
assert ngcs.list[0].end == 25
assert ngcs.list[1].start == 10
assert ngcs.list[1].end == 50
ngcs = genset.extend(mode="both", length=5)
assert ngcs.list[0].start == 5
assert ngcs.list[0].end == 25
assert ngcs.list[1].start == 10
assert ngcs.list[1].end == 55
def test_intersect_5():
"""
End-to-end attach
A : ------ ------
B : ------
R : none
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=11, end=20, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=5, end=11, name="test",
strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 0
res = genset1.intersect(genset2, mode="original")
assert len(res) == 0
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def test_intersect_4():
"""
No overlapping
A : ------ --------- -------
B : ---- ------ ------
R : none
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=11, end=20, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=33, end=38, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=7, end=9, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=20, end=25, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=26, end=31, name="test", strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 0
res = genset1.intersect(genset2, mode="original")
assert len(res) == 0
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def test_intersect_13():
"""
Completely included overlapping
A : ---------------------------
B : ---- ------ -----------
R1: ---- ------ ------ (overlap)
R2: --------------------------- (original)
R3: (comp_incl)
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=50, name="test",
strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=10, end=19, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=45, end=60, name="test", strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 3
res = genset1.intersect(genset2, mode="original")
assert len(res) == 1
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def test_intersect_15():
"""
A : -------------- -------
------
B : ----- ----------------
R1: ----- ------- (overlap)
----
R2: -------------- ------- (original)
------
R3: ------- (comp_incl)
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=50, name="test",
strand="."))
genset1.add(
GenCoor(chrom="chr1", start=20, end=40, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=70, end=80, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(
GenCoor(chrom="chr1", start=25, end=45, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=65, end=95, name="test", strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 3
res = genset1.intersect(genset2, mode="original")
assert len(res) == 3
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 1
def test_intersect_12():
"""
Different chromosomes
A : chr1 -------
B : chr2 -------
R : none
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=10, name="test",
strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr2", start=1, end=10, name="test",
strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 0
res = genset1.intersect(genset2, mode="original")
assert len(res) == 0
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def test_intersect_11():
"""
Many various overlapping (mixed)
A : ------------------ -------- ---------
B : ---- ------- ------ ----------
R1: -- ------- -- ---- --- (overlap)
R2: ------------------ -------- --------- (original)
R3: (comp_incl)
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=3, end=30, name="test",
strand="."))
genset1.add(
GenCoor(chrom="chr1", start=50, end=60, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=70, end=85, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=10, end=19, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=27, end=35, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=55, end=75, name="test", strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 5
res = genset1.intersect(genset2, mode="original")
assert len(res) == 3
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def test_intersect_6():
"""
No length attach
A : . .
B : . .
R : none
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=2, end=2, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=20, end=20, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=5, end=5, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=20, end=20, name="test", strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 1
res = genset1.intersect(genset2, mode="original")
assert len(res) == 1
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 1
if len(signals) > 1:
print("There are more than one BED files sharing the same tags, only the first one will be used.")
arrs = bed_sig2arr(beds[0], [signals[0]], exp, arg)
a = arrs[signals[0]]
hm = axes[i, j].imshow(a, cmap='hot', interpolation='None')
# Y ticks
axes[i, j].get_yaxis().set_ticks([])
# X ticks
x_label_list = ['-'+arg["--ext"], '0', arg["--ext"]]
xmin, xmax = axes[i, j].get_xlim()
axes[i, j].set_xticks([xmin, int(0.5 * (xmax - xmin)), xmax])
axes[i, j].set_xticklabels(x_label_list)
# labels
if j == 0:
axes[i, j].set_ylabel(row)
if i == 0:
regions = GenCoorSet(name=beds[0])
regions.load(filename=exp.get_file(beds[0]))
axes[i, j].set_title(col+" ("+str(len(regions))+")")
cbar_ax = fig.add_axes([0.9, 0.15, 0.02, 0.7])
fig.colorbar(hm, cax=cbar_ax)
set_yaxis(n_row, n_col, axes, arg)
fig.savefig(arg["<output_file>"], bbox_inches='tight')
# elif arg["boxplot"]:
def test_distances1():
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=5, name="test", strand="."))
genset1.add(
GenCoor(chrom="chr1", start=80, end=84, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(
GenCoor(chrom="chr1", start=10, end=15, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=60, end=64, name="test", strand="."))
genset2.add(
GenCoor(chrom="chr1", start=160, end=164, name="test", strand="."))
res = genset1.distances(genset2, sign=False)
print(res)
assert len(res) == 2
assert res == [5, 16]
coordinates.py split_strand <input_BED_file_path> <output_BED_directory_path>
Blablabla
Options:
"""
from docopt import docopt
from gencoor.coordinates import GenCoorSet
import os
if __name__ == '__main__':
arg = docopt(__doc__)
if arg["resize"]:
gc = GenCoorSet(name="input")
gc.load(arg["<input_BED_file_path>"], filetype="BED")
gc.relocate(mode='center as center', width=int(arg["<length>"]))
gc.save(arg["<output_BED_file_path>"], filetype="BED")
elif arg["split_strand"]:
name = os.path.basename(arg["<input_BED_file_path>"]).split(".")[0]
print(
os.path.join(arg["<output_BED_directory_path>"],
name + "_" + "+" + ".bed"))
gc = GenCoorSet(name="input")
gc.load(arg["<input_BED_file_path>"], filetype="BED")
res = gc.split_by_strands()
for k, g in res.items():
g.save(os.path.join(arg["<output_BED_directory_path>"],
name + "_" + k + ".bed"),
def test_intersect_8():
"""
One overlapping region
A : ------
B : --------
R1: -- (overlap)
R2: ------ (original)
R3: (comp_incl)
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=10, name="test",
strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=7, end=20, name="test",
strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 1
assert res[0].start == 7
assert res[0].end == 10
res = genset1.intersect(genset2, mode="original")
assert len(res) == 1
assert res[0].start == 1
assert res[0].end == 10
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def test_intersect_10():
"""
Two separately overlapping regions
A : ------- --------
B : ----- --------
R1: --- ---- (overlap)
R2: ------- -------- (original)
R3: (comp_incl)
"""
genset1 = GenCoorSet(name="Test_set")
genset1.add(GenCoor(chrom="chr1", start=1, end=10, name="test",
strand="."))
genset1.add(
GenCoor(chrom="chr1", start=26, end=35, name="test", strand="."))
genset2 = GenCoorSet(name="Test_set")
genset2.add(GenCoor(chrom="chr1", start=7, end=15, name="test",
strand="."))
genset2.add(
GenCoor(chrom="chr1", start=30, end=40, name="test", strand="."))
res = genset1.intersect(genset2, mode="overlap")
assert len(res) == 2
res = genset1.intersect(genset2, mode="original")
assert len(res) == 2
res = genset1.intersect(genset2, mode="complete_included")
assert len(res) == 0
def step_size(self):
return int(self.config["Parameters"]["step_size"])
def get_inputs(self):
res = {}
if "Input" in self.config.sections():
for lab in self.config["Input"]:
res[lab] = self.config["Input"][lab]
return res
if __name__ == '__main__':
arg = docopt(__doc__)
if arg["diffpeak"]:
config = DiffPeakConfig(filepath=arg["<config_file>"])
ref_back = GenCoorSet(name="background")
ref_back.get_chromosomes(genome=config.genome())
sig = SignalProfile(regions=ref_back,
genome=config.genome(),
bin=config.bin_size(),
step=config.step_size())
sig.load_files(file_dict=config.files_dict)
# Normalization
# sig.norm_bakcground(genome="hg38")
sig.norm_library_size()
# Input
if "Input" in config.config.sections():
sig2 = SignalProfile(regions=ref_back,
genome=config.genome(),