def test_sanitize_records_triu_action():
text = valid_data
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
out = sanitize_records(bins, schema="pairs", validate=True, tril_action="reflect")(
chunk.copy()
)
is_tril = ~np.array(out["triu"], dtype=bool)
is_tril_ix = out.index[is_tril]
assert np.all(out.loc[is_tril_ix, "chrom1"] == chunk.loc[is_tril_ix, "chrom2"])
assert np.all(out.loc[is_tril_ix, "chrom2"] == chunk.loc[is_tril_ix, "chrom1"])
assert np.all(out.loc[is_tril_ix, "strand1"] == "+")
text = valid_data
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
out = sanitize_records(bins, schema="pairs", validate=True, tril_action="drop")(
chunk.copy()
)
is_tril = ~np.array(out["triu"], dtype=bool)
is_tril_ix = out.index[is_tril]
assert np.all(out.loc[is_tril_ix, "chrom1"] == chunk.loc[is_tril_ix, "chrom2"])
assert np.all(out.loc[is_tril_ix, "chrom2"] == chunk.loc[is_tril_ix, "chrom1"])
assert np.all(out.loc[is_tril_ix, "strand1"] == "+")
assert len(out) == chunk["triu"].sum()
func = sanitize_records(bins, schema="pairs", validate=True, tril_action="raise")
text = valid_data
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
with pytest.raises(BadInputError):
func(chunk)
def test_aggregate_records():
bins = cooler.binnify(
cooler.util.read_chromsizes(op.join(datadir, "toy.chrom.sizes")), 1
)
records = pd.read_csv(
op.join(datadir, "toy.pairs"),
sep='\t',
names=[
"read_id",
"chrom1", "pos1",
"chrom2", "pos2",
"strand1", "strand2",
"value"
]
)
sanitizer = sanitize_records(
bins,
schema="pairs",
validate=False,
tril_action="reflect",
is_one_based=False,
sort=False,
)
chunk = sanitizer(records)
aggregator = aggregate_records()
aggregator(chunk)
def test_sanitize_records_with_nuisance_records():
text = _insert_lines(valid_data, nuisance_chroms)
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
out = sanitize_records(bins, schema="pairs", validate=True, tril_action="reflect")(
chunk.copy()
)
assert ("chr9" not in out["chrom1"]) and ("chr9" not in out["chrom2"])
def test_sanitize_triu_action():
text = valid_data
chunk = pd.read_csv(StringIO(text), sep='\t', names=columns)
out = sanitize_records(
bins,
schema='pairs',
validate=True,
tril_action='reflect',
)(chunk.copy())
is_tril = ~np.array(out['triu'], dtype=bool)
is_tril_ix = out.index[is_tril]
assert np.all(out.loc[is_tril_ix, 'chrom1'] == chunk.loc[is_tril_ix,
'chrom2'])
assert np.all(out.loc[is_tril_ix, 'chrom2'] == chunk.loc[is_tril_ix,
'chrom1'])
assert np.all(out.loc[is_tril_ix, 'strand1'] == '+')
text = valid_data
chunk = pd.read_csv(StringIO(text), sep='\t', names=columns)
out = sanitize_records(
bins,
schema='pairs',
validate=True,
tril_action='drop',
)(chunk.copy())
is_tril = ~np.array(out['triu'], dtype=bool)
is_tril_ix = out.index[is_tril]
assert np.all(out.loc[is_tril_ix, 'chrom1'] == chunk.loc[is_tril_ix,
'chrom2'])
assert np.all(out.loc[is_tril_ix, 'chrom2'] == chunk.loc[is_tril_ix,
'chrom1'])
assert np.all(out.loc[is_tril_ix, 'strand1'] == '+')
assert len(out) == chunk['triu'].sum()
func = sanitize_records(
bins,
schema='pairs',
validate=True,
tril_action='raise',
)
text = valid_data
chunk = pd.read_csv(StringIO(text), sep='\t', names=columns)
with pytest.raises(BadInputError):
func(chunk)
def test_sanitize_with_nuisance_records():
text = _insert_lines(valid_data, nuisance_chroms)
chunk = pd.read_csv(StringIO(text), sep='\t', names=columns)
out = sanitize_records(
bins,
schema='pairs',
validate=True,
tril_action='reflect',
)(chunk.copy())
assert ('chr9' not in out['chrom1']) and ('chr9' not in out['chrom2'])
def test_sanitize_records_with_bad_records():
func = sanitize_records(bins, schema="pairs", validate=True, tril_action="reflect")
text = _insert_lines(valid_data, oob_lower)
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
with pytest.raises(BadInputError):
func(chunk)
text = _insert_lines(valid_data, oob_upper)
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
with pytest.raises(BadInputError):
func(chunk)
def test_sanitize_with_strand_column():
text = valid_data
chunk = pd.read_csv(StringIO(text), sep='\t', names=columns)
out = sanitize_records(
bins,
schema='pairs',
validate=True,
tril_action='reflect',
sided_fields=('chrom', 'pos', 'strand'),
)(chunk.copy())
is_tril = ~np.array(out['triu'], dtype=bool)
assert np.all(out.loc[is_tril, 'chrom1'] == chunk.loc[is_tril, 'chrom2'])
assert np.all(out.loc[is_tril, 'chrom2'] == chunk.loc[is_tril, 'chrom1'])
assert np.all(out.loc[is_tril, 'strand1'] == '-')
def test_sanitize_records_with_strand_column():
text = valid_data
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
out = sanitize_records(
bins,
schema="pairs",
validate=True,
tril_action="reflect",
sided_fields=("chrom", "pos", "strand"),
)(chunk.copy())
is_tril = ~np.array(out["triu"], dtype=bool)
assert np.all(out.loc[is_tril, "chrom1"] == chunk.loc[is_tril, "chrom2"])
assert np.all(out.loc[is_tril, "chrom2"] == chunk.loc[is_tril, "chrom1"])
assert np.all(out.loc[is_tril, "strand1"] == "-")
def test_sanitize_with_bad_records():
func = sanitize_records(
bins,
schema='pairs',
validate=True,
tril_action='reflect',
)
text = _insert_lines(valid_data, oob_lower)
chunk = pd.read_table(StringIO(text), names=columns)
with pytest.raises(BadInputError):
func(chunk)
text = _insert_lines(valid_data, oob_upper)
chunk = pd.read_table(StringIO(text), names=columns)
with pytest.raises(BadInputError):
func(chunk)
def test_sanitize_records():
chunk = pd.read_csv(StringIO(valid_data), sep="\t", names=columns)
with pytest.raises(ValueError):
sanitize_records(
bins,
schema="doesnotexist",
validate=True,
tril_action="reflect",
is_one_based=True,
sort=True,
)(chunk.copy())
chunk = pd.read_csv(StringIO(valid_data), sep="\t", names=columns)
sanitize_records(
bins,
schema="pairs",
validate=True,
tril_action="reflect",
is_one_based=True,
sort=True,
)(chunk.copy())
# variable-length bins
chunk = pd.read_csv(StringIO(valid_data), sep="\t", names=columns)
sanitize_records(
pd.DataFrame({
'chrom': ['chr1', 'chr1', 'chr2', 'chr2', 'chr3'],
'start': [0, 150, 0, 100, 0],
'end': [150, 300, 100, 300, 300],
}),
schema="pairs",
validate=True,
tril_action="reflect",
is_one_based=True,
sort=True,
)(chunk.copy())
# input with already enum-encoded chromosomes (decode_chroms=False)
text = """0\t1\t+\t1\t100\t-\t.\tLL\t1
1\t99\t+\t0\t13\t-\t.\tLL\t0
1\t13\t+\t1\t60\t-\t.\tLL\t1
0\t200\t+\t1\t50\t-\t.\tLL\t1
2\t11\t+\t2\t40\t-\t.\tLL\t1
0\t234\t+\t2\t30\t-\t.\tLL\t1
2\t3\t+\t1\t20\t-\t.\tLL\t0
1\t23\t+\t2\t11\t-\t.\tLL\t1
0\t123\t+\t-1\t200\t-\t.\tLL\t1
"""
chunk = pd.read_csv(StringIO(text), sep="\t", names=columns)
sanitize_records(
bins,
schema="pairs",
decode_chroms=False,
validate=True,
tril_action="reflect"
)(chunk.copy())
# fails on string chromosomes
chunk = pd.read_csv(StringIO(valid_data), sep="\t", names=columns)
with pytest.raises(BadInputError):
sanitize_records(
bins,
schema="pairs",
decode_chroms=False,
validate=True,
tril_action="reflect"
)(chunk.copy())
# empty chunk
out = sanitize_records(
bins,
schema="pairs",
validate=True,
tril_action="reflect"
)(chunk.iloc[0:0])
assert len(out) == 0
