def test_labelledallele_delabeler():
ngenos = 10 # Number of genotypes per chromosome
if ngenos % 2 == 1:
raise ValueError('Even number of genotypes needed')
p = Population()
c = ChromosomeTemplate()
for i in range(ngenos):
c.add_genotype()
p.add_chromosome(c)
a = Individual(p, 1)
a._init_genotypes(blankchroms=False)
a.genotypes[0][0] = Alleles([1]*ngenos)
a.genotypes[0][1] = Alleles([2]*ngenos)
b = Individual(p, 2)
b._init_genotypes(blankchroms=False)
b.genotypes[0][0] = Alleles([3] * ngenos)
b.genotypes[0][1] = Alleles([4] * ngenos)
chromatid_spans = [InheritanceSpan(a, 0, 0, 0, ngenos//2),
InheritanceSpan(b, 0, 1, ngenos//2, ngenos)]
chromatid = LabelledAlleles(spans=chromatid_spans, chromobj=c)
expected_value = [1]*(ngenos//2) + [4] * (ngenos//2)
expected_value = Alleles(expected_value)
actual_value = chromatid.delabel()
assert all(actual_value == expected_value)
def test_labelledallele_delabeler():
ngenos = 10 # Number of genotypes per chromosome
if ngenos % 2 == 1:
raise ValueError('Even number of genotypes needed')
p = Population()
c = ChromosomeTemplate()
for i in range(ngenos):
c.add_genotype()
p.add_chromosome(c)
a = Individual(p, 1)
a._init_genotypes(blankchroms=False)
a.genotypes[0][0] = Alleles([1] * ngenos)
a.genotypes[0][1] = Alleles([2] * ngenos)
b = Individual(p, 2)
b._init_genotypes(blankchroms=False)
b.genotypes[0][0] = Alleles([3] * ngenos)
b.genotypes[0][1] = Alleles([4] * ngenos)
chromatid_spans = [
InheritanceSpan(a, 0, 0, 0, ngenos // 2),
InheritanceSpan(b, 0, 1, ngenos // 2, ngenos)
]
chromatid = LabelledAlleles(spans=chromatid_spans, chromobj=c)
expected_value = [1] * (ngenos // 2) + [4] * (ngenos // 2)
expected_value = Alleles(expected_value)
actual_value = chromatid.delabel()
assert all(actual_value == expected_value)
def test_labelledalleles():
IS = InheritanceSpan
ngenos = 50
p = Population()
c = ChromosomeTemplate()
for i in range(ngenos):
c.add_genotype()
p.add_chromosome(c)
a = Individual(p, 1)
actual = LabelledAlleles.founder_chromosome(a, 0, 0, chromobj=c)
expected = LabelledAlleles(spans=[IS(a, 0, 0, 0, ngenos)], chromobj=c)
assert actual == expected
def test_labelledalleles():
IS = InheritanceSpan
ngenos = 50
p = Population()
c = ChromosomeTemplate()
for i in range(ngenos):
c.add_genotype()
p.add_chromosome(c)
a = Individual(p, 1)
actual = LabelledAlleles.founder_chromosome(a, 0, 0, chromobj=c)
expected = LabelledAlleles(spans=[IS(a, 0, 0, 0, ngenos)], chromobj=c)
assert actual == expected
def read_vcf(filename, require_pass=False, freq_info=None, info_filters=None):
'''
Reads a VCF file and returns a Population object with the
individuals represented in the file
'''
if not info_filters:
info_filters = []
for filter in info_filters:
if not callable(filter):
raise ValueError('Filter not callable')
with open(filename) as f:
pop = Population()
last_chrom = None
genotypes = []
for i, line in enumerate(f):
if line.startswith('##'):
continue
elif line.startswith('#'):
ind_ids = line.strip().split()[9:]
inds = [Individual(pop, ind_id) for ind_id in ind_ids]
for ind in inds:
pop.register_individual(ind)
break
for i, line in enumerate(f):
record = VCFRecord(line)
if info_filters and not all(filter(record) for filter in info_filters):
continue
if require_pass and not record.filter_passed:
continue
if record.chrom != last_chrom:
if last_chrom is not None:
chromobj.finalize()
pop.add_chromosome(chromobj)
chromobj = ChromosomeTemplate(label=record.chrom)
if freq_info is not None and freq_info in record.info:
freq = record.info[freq_info]
if ',' in freq:
freq = freq.split(',')[0]
freq = float(freq)
else:
freq = 0
genorow = record.genotypes()
genotypes.append(genorow)
chromobj.add_genotype(bp=record.pos,
label=record.label,
frequency=freq)
last_chrom = record.chrom
chromobj.finalize()
pop.add_chromosome(chromobj)
for ind in inds:
# Initialize new genotypes
ind._init_genotypes(sparse=True)
# Now actually sift through markers and assign them to individuals
final_indices = []
for chromidx, chromobj in enumerate(pop.chromosomes):
indices = zip([chromidx]*chromobj.nmark(), range(chromobj.nmark()))
final_indices.extend(indices)
raw_indices = range(len(genotypes))
for raw, final in zip(raw_indices, final_indices):
chromidx, markidx = final
row = genotypes[raw]
assign_genorow(row, inds, chromidx, markidx)
# Kill the row so we don't end up with the whole dataset in memory twice
genotypes[raw] = None
return pop
