def test_children_len():
x = Locus('1',1,2)
y = Locus('1',3,4,name='sublocus1')
z = Locus('1',3,4,name='sublocus2')
x.children = (y,z)
assert len(x.children) == 2
def test_combine():
x = Locus('1',1,2)
y = Locus('1',3,4)
z = x.combine(y)
assert z.start == 1
assert z.end == 4
assert len(z.children) == 2
def test_add_duplicate_locus():
t = Term('test')
assert len(t) == 0
t.add_locus(Locus('1', 1, 100))
assert len(t) == 1
# Add duplicate which wont count
t.add_locus(Locus('1', 1, 100))
assert len(t) == 1
def simpleRefLoci():
# Create a Locus
a = Locus(1,100,150, id='gene_a')
# Create a couple more!
b = Locus(1,160,175, id='gene_b')
c = Locus(1,180,200, id='gene_c')
d = Locus(1,210,300, id='gene_d')
e = Locus(2,100,150, id='gene_e')
x = RefLoci('simpleRefLoci')
x.add_loci([a,b,c,d,e])
return x
def t():
a = Locus('1', 1, 100)
b = Locus('2', 200, 300)
c = Locus('3', 300, 400)
t = Term('test',
desc='test term',
loci=[a, b, c],
attrs={
'foo': 'bar',
'baz': 'bat'
})
return t
def testRefGen():
# Create a Locus
a = Locus(8,100,150, name='gene_a')
# Createa couple more!
b = Locus(8,160,175, name='gene_b')
c = Locus(8,180,200, name='gene_c')
d = Locus(8,210,300, name='gene_d')
e = Locus(9,100,150, name='gene_e')
x = Loci()
x.append([a,b,c,d,e])
return x
def test_nuke_tables():
'add a locus to an empty refloci db and then retrieve it'
if m80.Tools.available('RefLoci', 'empty'):
m80.Tools.delete('RefLoci', 'empty', force=True)
empty = RefLoci('empty')
assert len(empty) == 0
x = Locus('1', 1, 1, feature_type='gene', attrs={'foo': 'bar'})
y = Locus('1', 2, 2, feature_type='exon', attrs={'baz': 'bat'})
x.add_sublocus(y)
LID = empty.add_locus(x)
assert len(empty) == 1
empty._nuke_tables()
assert len(empty) == 0
m80.Tools.delete('RefLoci', 'empty', force=True)
def test_add_locus_with_subloci():
'add a locus to an empty refloci db and then retrieve it'
if m80.Tools.available('RefLoci', 'empty'):
m80.Tools.delete('RefLoci', 'empty', force=True)
empty = RefLoci('empty')
assert len(empty) == 0
x = Locus('1', 1, 1, feature_type='gene', attrs={'foo': 'bar'})
y = Locus('1', 2, 2, feature_type='exon', attrs={'baz': 'bat'})
x.add_sublocus(y)
LID = empty.add_locus(x)
assert len(empty) == 1
l = empty._get_locus_by_LID(LID)
assert l['foo'] == 'bar'
assert len(l.subloci) == 1
m80.Tools.delete('RefLoci', 'empty', force=True)
def test_loci_within(testRefGen):
random_locus = testRefGen.random_locus()
bigger_locus = Locus(random_locus.chrom,
start=random_locus.start - 100,
end=random_locus.end + 100)
genes = testRefGen.genes_within(bigger_locus)
assert random_locus in genes
def test_get_attrs(simpleRefLoci):
nLocus = Locus(1, 2, 3, 'nLocus', foo="bar", baz="bat")
simpleRefLoci.add_locus(nLocus)
del nLocus
nLocus = simpleRefLoci['nLocus']
assert nLocus['foo'] == 'bar'
assert nLocus['baz'] == 'bat'
def test_removie_locus(simpleRefLoci):
new_locus = Locus(1, 100, 300, id='new_locus')
if new_locus not in simpleRefLoci:
simpleRefLoci.add_locus(new_locus)
assert new_locus in simpleRefLoci
simpleRefLoci.remove_locus(new_locus.id)
assert new_locus not in simpleRefLoci
def test_within_bad_strand(testRefGen):
'simple within to get chromosomal segment'
with pytest.raises(StrandError):
assert len(
list(
testRefGen.within(Locus('1', 1, 139000, strand='='),
partial=False))) == 4
def test_within_error_on_both_same_strand_and_ignore_strand(testRefGen):
try:
testRefGen.within(Locus('1', 1, 139000, strand='-'),
ignore_strand=True,
same_strand=True)
except ValueError as e:
assert True
def test_upstream_plus_strand(testRefGen):
# Below is GRMZM2G093399, but on the minus strand
x = Locus('1', 136307, 138929)
l = [x.name for x in testRefGen.upstream_loci(x, n=3)]
assert l[0] == 'GRMZM2G093344'
assert l[1] == 'GRMZM5G888250'
assert l[2] == 'GRMZM2G059865'
def test_within_same_strand_minus(testRefGen):
'test fetching loci only on the same strand'
assert len(
list(
testRefGen.within(Locus('1', 1, 139000, strand='-'),
partial=True,
same_strand=True))) == 3
def test_within_same_strand_and_ignore_strand(testRefGen):
'test fetching loci only on the same strand'
with pytest.raises(ValueError):
list(
testRefGen.within(Locus('1', 1, 139000, strand='+'),
ignore_strand=True,
same_strand=True))
def test_downstream(testRefGen):
l = [
x.name for x in testRefGen.downstream_loci(Locus('1', 4854, 9652), n=3)
]
assert l[0] == 'GRMZM5G888250'
assert l[1] == 'GRMZM2G093344'
assert l[2] == 'GRMZM2G093399'
def test_downstream_same_strand_limit_n(testRefGen):
l = [
x.name for x in testRefGen.downstream_loci(
Locus('1', 4854, 9652), n=1, same_strand=True)
]
assert len(l) == 1
# This skips GRMZM5G888250 and GRMZM2G093344 since they are on the - strand
assert l[0] == 'GRMZM2G093399'
def test_add_locus():
'add a locus to an empty refloci db and then retrieve it'
if m80.Tools.available('RefLoci', 'empty'):
m80.Tools.delete('RefLoci', 'empty', force=True)
empty = RefLoci('empty')
assert len(empty) == 0
empty.add_locus(Locus('1', 1, 1, feature_type='gene', attrs={'foo':
'bar'}))
assert len(empty) == 1
m80.Tools.delete('RefLoci', 'empty', force=True)
def test_flank_limit_for_candidate_loci_from_SNP(testRefGen):
random_locus = testRefGen.random_locus()
downstream = testRefGen.downstream_loci(random_locus,
locus_limit=10,
window_size=50e6)
test_snp = Locus(downstream[5].chrom, downstream[5].start, window=50e6)
# Create a Locus that is on gene 5
candidates = testRefGen.candidate_loci(test_snp,
flank_limit=5,
window_size=50e6)
assert len(candidates) == 11
def test_candidate_loci_from_SNP(testRefGen):
random_locus = testRefGen.random_locus()
# grab a bunch of downstream genes
down1, down2 = testRefGen.downstream_loci(random_locus,
locus_limit=2,
window_size=50e6)
# Create a Locus containing both down1 and down2
test_snp = Locus(down1.chrom,
down1.start - 50,
end=down2.end + 50,
window=50e6)
candidates = testRefGen.candidate_loci(test_snp,
flank_limit=5,
chain=False)
assert len(candidates) == 12
def test_children_detach_in_children_setter():
x = Locus('1',1,2)
y = Locus('1',3,4,name='sublocus1')
z = Locus('1',3,4,name='sublocus2')
x.children = [y]
# Now trigger a detach for current children
# when assigning new children
x.children = [z]
assert len(x.children) == 1
def __getitem__(self, item):
'''
A convenience method to extract loci
by their name attribute
e.g.
# Create a locus object
>>> x = Locus('chr1',1,100,name='locus1')
>>> y = Loci['locus1']
>>> x == y
True
'''
try:
LID, = self._db.cursor().execute(
'''
SELECT LID FROM loci WHERE name = ?
''', (item, )).fetchone()
except TypeError:
raise ValueError(f'{item} not in Loci')
from locuspocus import Locus
return Locus.from_LID(LID)
def _get_locus_by_LID(self, LID: int):
'''
Get a locus by its LID
Parameters
----------
LID : int
A Locus ID. These are assigned to Locus objects when
they are added to the Loci database.
Returns
-------
The Locus.
Raises
------
`MissingLocusError` if there is no Locus in the database with that LID.
'''
lid_exists, = self._db.cursor().execute(
'SELECT COUNT(*) FROM loci WHERE LID = ? ', (LID, )).fetchone()
if lid_exists == 0:
raise MissingLocusError
from locuspocus import Locus
return Locus(LID, self)
def test_from_record(simple_Locus):
assert Locus.from_record(
('1', 100, 200, '<None>1:100-200', 0, '<None>1:100-200')) == simple_Locus
def test_get_LID_missing(testRefGen):
'Make sure that fetching a locus by LID returns the same locus'
with pytest.raises(MissingLocusError):
assert testRefGen._get_LID(Locus('na', 1, 1))
def test_encompassing_loci(testRefGen):
x = Locus('1', 10000, 10000)
loci = list(testRefGen.encompassing_loci(x))
assert loci[0].name == 'GRMZM5G888250'
def test_flank_loci_limited_n(testRefGen):
x = Locus('1', 10500, 10500)
up, down = testRefGen.flanking_loci(x, n=2)
assert len(list(up)) == 2
assert len(list(down)) == 2
def test_within_strand_order_minus(testRefGen):
# should return fourth gene first
loci = list(testRefGen.within(Locus('1', 1, 139000, strand='-')))
assert loci[0].start == 136307
def test_downstream_limit_n(testRefGen):
l = [
x.name for x in testRefGen.downstream_loci(Locus('1', 4854, 9652), n=1)
]
assert l[0] == 'GRMZM5G888250'
assert len(l) == 1
def test_downstream_accepts_loci(testRefGen):
x = Locus('1', 4854, 9652)
loci = [x, x]
l1, l2 = map(list, testRefGen.downstream_loci(loci, n=2))
assert len(l1) == 2
assert len(l2) == 2
