class TestTrioModel(unittest.TestCase):
def setUp(self):
self.trio_model = TrioModel(reads=[[30, 0, 0, 0], [30, 0, 0, 0], [30, 0, 0, 0]])
def test_pop_sample(self):
parent_prob_mat = self.trio_model.pop_sample()
proba = np.sum(parent_prob_mat)
self.assertAlmostEqual(proba, 1)
def test_germ_muta(self):
proba = np.sum(self.trio_model.child_prob_mat)
self.assertAlmostEqual(proba, 256)
def test_soma_muta(self):
proba = np.sum(self.trio_model.soma_prob_mat)
self.assertAlmostEqual(proba, 16)
def test_seq_err(self):
seq_prob_mat = self.trio_model.seq_err_all()
proba = np.sum(seq_prob_mat)
self.assertAlmostEqual(proba, 3)
def test_trio(self):
proba = self.trio_model.trio()
self.assertAlmostEqual(proba, 0)
def __init__(self, coverage, germ_muta_rate, soma_muta_rate):
"""
Generate a random sample and calculate probability of mutation with the
given coverage.
Coverage, and germline and somatic mutation rates are adjustable via
command line.
"""
if germ_muta_rate is not None and soma_muta_rate is not None:
self.trio_model = TrioModel(germ_muta_rate=germ_muta_rate,
soma_muta_rate=soma_muta_rate)
elif germ_muta_rate is not None:
self.trio_model = TrioModel(germ_muta_rate=germ_muta_rate)
elif soma_muta_rate is not None:
self.trio_model = TrioModel(soma_muta_rate=soma_muta_rate)
else:
self.trio_model = TrioModel()
self.cov = coverage
self.has_muta = False
def write_proba(child, mother, father, filename):
"""
Write the probability of each site on a new line to a text file.
Args:
child: Iterable of the content in the child pileup file.
mother: Iterable of the content in the mother pileup file.
father: Iterable of the content in the father pileup file.
filename: String representing the name of the output file.
"""
child_lines = trim_header(child)
mother_lines = trim_header(mother)
father_lines = trim_header(father)
fout = open(filename, 'w')
for c, m, f in itertools.zip_longest(child_lines, mother_lines, father_lines):
reads = [get_reads(c), get_reads(m), get_reads(f)]
trio_model = TrioModel(reads=reads)
proba = trio_model.trio()
if proba >= THRESHOLD:
site_proba = '%s\t%s\n' % (c.split("\t")[1], str(proba))
fout.write(site_proba)
fout.close()
def write_proba(child, mother, father, filename):
"""
Write the probability of each site on a new line to a text file.
Args:
child: Iterable of the content in the child pileup file.
mother: Iterable of the content in the mother pileup file.
father: Iterable of the content in the father pileup file.
filename: String representing the name of the output file.
"""
child_lines = trim_header(child)
mother_lines = trim_header(mother)
father_lines = trim_header(father)
fout = open(filename, 'w')
for c, m, f in itertools.zip_longest(child_lines, mother_lines,
father_lines):
reads = [get_reads(c), get_reads(m), get_reads(f)]
trio_model = TrioModel(reads=reads)
proba = trio_model.trio()
if proba >= THRESHOLD:
site_proba = '%s\t%s\n' % (c.split("\t")[1], str(proba))
fout.write(site_proba)
fout.close()
def setUp(self):
self.trio_model = TrioModel(reads=[[30, 0, 0, 0], [30, 0, 0, 0], [30, 0, 0, 0]])
for line in handle:
values = line.strip('\n').split('\t')
child_read_arr = values[:4]
mom_read_arr = values[4:8]
dad_read_arr = values[8:12]
child_read = [int(count) for count in child_read_arr]
mom_read = [int(count) for count in mom_read_arr]
dad_read = [int(count) for count in dad_read_arr]
reads = [child_read, mom_read, dad_read]
rates_arr = values[12:16]
rates = [float(rate) for rate in rates_arr]
disp = float(values[16]) if values[16] else 1000 # default dispersion value
bias = float(values[17]) if values[17] else None
trio_model = TrioModel(
reads=reads,
pop_muta_rate=rates[0],
germ_muta_rate=rates[1],
soma_muta_rate=rates[2],
seq_err_rate=rates[3],
dm_disp=disp,
dm_bias=bias
)
proba = trio_model.trio()
print(reads, end='\t')
print(proba)
handle.close()
# run python driver.py <parameters.txt>
handle = open(sys.argv[1])
for line in handle:
values = line.strip('\n').split('\t')
child_read_arr = values[:4]
mom_read_arr = values[4:8]
dad_read_arr = values[8:12]
child_read = [int(count) for count in child_read_arr]
mom_read = [int(count) for count in mom_read_arr]
dad_read = [int(count) for count in dad_read_arr]
reads = [child_read, mom_read, dad_read]
rates_arr = values[12:16]
rates = [float(rate) for rate in rates_arr]
disp = float(values[16]) if values[16] else 1000 # default dispersion value
bias = float(values[17]) if values[17] else None
trio_model = TrioModel(
reads=reads,
pop_muta_rate=rates[0],
germ_muta_rate=rates[1],
soma_muta_rate=rates[2],
seq_err_rate=rates[3],
dm_disp=disp,
dm_bias=bias
)
proba = trio_model.trio()
print(reads, end='\t')
print(proba)
handle.close()