def test_select_samples(self):
'''Create sub-problem by selecting samples/family..'''
p = io.read_npz(itu.FAMILY13_STAGE3)
f = p.families()[0]
ps = p.sub_problem(f.member_list)
pf = p.sub_problem_of_parents(f.father, f.mother)
# pf has family_info entry, ps doesn't
assert_equal(ps.pedigree.graph.nodes(), pf.pedigree.graph.nodes(),
'Incorrect pedigree graph nodes')
assert_equal(set(ps.pedigree.graph.edges()),
set(pf.pedigree.graph.edges()),
'Incorrect pedigree graph edges')
assert_equal(ps.pedigree.sample_id.tolist(),
pf.pedigree.sample_id.tolist(),
'Incorrect pedigree sample ID array')
assert_equal(ps.pedigree, pf.pedigree, 'Incorrect pedigree')
assert_equal(ps.genotype, pf.genotype, 'Incorrect genotype')
assert_equal(ps.haplotype, pf.haplotype, 'Incorrect haplotype')
self.assertTrue(ps.info.family_info != pf.info.family_info,
'Incorrect info family')
assert_not_equal(ps.info, pf.info, 'Incorrect info')
assert_not_equal(ps, pf,
'pf should have a family_info entry, ps shouldn'
't')
def __load_problem(options):
'''Load data from input files into a Problem object.'''
# args = dict(prefix=options.prefix, tped=options.tped, tfam=options.tfam)
# print args
if options.debug: print 'Input file : %s ...\nOutput file: %s ...' % (options.input, options.output,)
problem = io.read_npz(options.input)
if options.debug: print 'Loaded', problem
return problem
def __save_and_load_problem_npz(self, problem):
'''Save and load a problem from NPZ file.'''
out_file = tempfile.TemporaryFile()
io.write_npz(problem, out_file)
# Only needed here to simulate closing & reopening file; you will need to call
# out_file.close() on to prevent file locking in Windows
out_file.seek(0)
return io.read_npz(out_file)
def test_global_ibd_segments(self):
'''Test the construction of the global IBD segment dictionary.'''
problem = io.read_npz(itu.FAMILY225_STAGE1)
ibd = problem.info.ibd
assert_equal(ibd.length, 0, 'Unexpected initial # of IBD segments')
self.phaser.run(problem)
assert_equal(ibd.length, 21, 'Unexpected final # of IBD segments before grouping')
ibd.group_to_disjoint()
assert_equal(ibd.length, 32, 'Unexpected final # of IBD segments after grouping')
def test_child_comparison_one_parent(self):
'''Test applying child comparison to a nuclear family with many genotyped kids but only
one genotyped parent.'''
problem = io.read_npz(itu.FAMILY945_ONE_PARENT_STAGE2)
itu.assert_size_equals(problem.genotype, 3218, 8)
itu.assert_problem_stats(problem, 51488, 44150, 96)
phaser = family_child_comparison_phaser(debug=False)
phaser.run(problem)
itu.assert_problem_stats(problem, 51488, 47343, 101)
def test_child_comparison_one_parent(self):
'''Test applying child comparison to a nuclear family with many genotyped kids but only
one genotyped parent.'''
problem = io.read_npz(itu.FAMILY945_ONE_PARENT_STAGE2)
itu.assert_size_equals(problem.genotype, 3218, 8)
itu.assert_problem_stats(problem, 51488, 44150, 96)
phaser = family_child_comparison_phaser(debug=False)
phaser.run(problem)
itu.assert_problem_stats(problem, 51488, 47343, 101)
def __plink_to_npz(self, p, file_name):
'''Convert p from plink to npz format using the file set specified by file_name.'''
npz = file_name+'.npz'
# Save test problem in plink format
io.write_plink(p, prefix=file_name)
# Convert plink -> npz
io.plink_to_npz(file_name, npz)
# Load npz and check that the problem object didn't change
p2 = io.read_npz(npz)
return p2
def test_family_12(self):
'''Test comparing sibs with non-genotyped parents (stage 4).'''
problem = io.read_npz(itu.FAMILY12_STAGE2)
itu.assert_size_equals(problem.genotype, 3218, 7)
itu.assert_problem_stats(problem, 45052, 42162, 237)
assert_equal(len(list(problem.families(genotyped=False))), 1, 'Incorrect number of families')
phaser = family_sib_comparison_phaser()
phaser.run(problem, PhaseParam(single_member=1))
itu.assert_problem_stats(problem, 45052, 42162, 237)
def test_family_963(self):
'''Test comparing sibs with non-genotyped parents (stage 4). This was a problematic family.'''
problem = io.read_npz(itu.FAMILY963_STAGE4)
itu.assert_size_equals(problem.genotype, 3218, 3)
itu.assert_problem_stats(problem, 19308, 19286, 23)
assert_equal(len(list(problem.families(genotyped=False))), 1, 'Incorrect number of families')
phaser = family_sib_comparison_phaser()
phaser.run(problem)
itu.assert_problem_stats(problem, 19308, 19286, 23)
def setUp(self):
'''Load test data and expected results.'''
unittest.TestCase.setUp(self)
# Load test data ready from previous phasing stagees
self.problem = io.read_npz(itu.FAMILY_TOO_ZEROED_STAGE1)
self.family = self.problem.families()[0]
self.phaser = phase_core.PhaseDecorator(FilterChain([trivial_phaser(),
family_phaser(),
family_child_comparison_phaser()]))
self.comparator = ic.ChildComparator(Struct(problem=self.problem, params=PhaseParam()), self.family)
def __load_problem(options):
'''Load data from input files into a Problem object.'''
# args = dict(prefix=options.prefix, tped=options.tped, tfam=options.tfam)
# print args
if options.debug:
print 'Input file : %s ...\nOutput file: %s ...' % (
options.input,
options.output,
)
problem = io.read_npz(options.input)
if options.debug: print 'Loaded', problem
return problem
def test_family_12(self):
'''Test comparing sibs with non-genotyped parents (stage 4).'''
problem = io.read_npz(itu.FAMILY12_STAGE2)
itu.assert_size_equals(problem.genotype, 3218, 7)
itu.assert_problem_stats(problem, 45052, 42162, 237)
assert_equal(len(list(problem.families(genotyped=False))), 1,
'Incorrect number of families')
phaser = family_sib_comparison_phaser()
phaser.run(problem, PhaseParam(single_member=1))
itu.assert_problem_stats(problem, 45052, 42162, 237)
def test_family_963(self):
'''Test comparing sibs with non-genotyped parents (stage 4). This was a problematic family.'''
problem = io.read_npz(itu.FAMILY963_STAGE4)
itu.assert_size_equals(problem.genotype, 3218, 3)
itu.assert_problem_stats(problem, 19308, 19286, 23)
assert_equal(len(list(problem.families(genotyped=False))), 1,
'Incorrect number of families')
phaser = family_sib_comparison_phaser()
phaser.run(problem)
itu.assert_problem_stats(problem, 19308, 19286, 23)
def setUp(self):
'''Load test data and expected results.'''
unittest.TestCase.setUp(self)
# Load test data ready from previous phasing stagees
self.problem = io.read_npz(itu.FAMILY13_STAGE2)
self.family = self.problem.families()[0]
self.phaser = phase_core.PhaseDecorator(
FilterChain([
trivial_phaser(),
family_phaser(),
family_child_comparison_phaser()
]))
self.comparator = ic.ChildComparator(
Struct(problem=self.problem, params=PhaseParam()), self.family)
def test_family_2003_need_poo_alignment(self):
'''Test comparing sibs with non-genotyped parents (stage 4). This case highlights
the need to align POO-phases, i.e., swap founder haps to correctly patch families at individual
ID 28412 (our original index 386; in this problem, index 10).'''
problem = io.read_npz(itu.FAMILY2003_STAGE3)
itu.assert_size_equals(problem.genotype, 3218, 9)
itu.assert_problem_stats(problem, 57924, 43339, 85)
assert_equal(len(list(problem.families(genotyped=False))), 1, 'Incorrect number of families')
#f = problem.families(genotyped=False)[0]
#print f.member_list
#print problem.pedigree.sample_id
phaser = family_sib_comparison_phaser()
phaser.run(problem)
itu.assert_problem_stats(problem, 57924, 57515, 85)
def test_family_2003_need_poo_alignment(self):
'''Test comparing sibs with non-genotyped parents (stage 4). This case highlights
the need to align POO-phases, i.e., swap founder haps to correctly patch families at individual
ID 28412 (our original index 386; in this problem, index 10).'''
problem = io.read_npz(itu.FAMILY2003_STAGE3)
itu.assert_size_equals(problem.genotype, 3218, 9)
itu.assert_problem_stats(problem, 57924, 43339, 85)
assert_equal(len(list(problem.families(genotyped=False))), 1,
'Incorrect number of families')
#f = problem.families(genotyped=False)[0]
#print f.member_list
#print problem.pedigree.sample_id
phaser = family_sib_comparison_phaser()
phaser.run(problem)
itu.assert_problem_stats(problem, 57924, 57515, 85)
def test_select_samples(self):
'''Create sub-problem by selecting samples/family..'''
p = io.read_npz(itu.FAMILY13_STAGE3)
f = p.families()[0]
ps = p.sub_problem(f.member_list)
pf = p.sub_problem_of_parents(f.father, f.mother)
# pf has family_info entry, ps doesn't
assert_equal(ps.pedigree.graph.nodes(), pf.pedigree.graph.nodes(), 'Incorrect pedigree graph nodes')
assert_equal(set(ps.pedigree.graph.edges()), set(pf.pedigree.graph.edges()), 'Incorrect pedigree graph edges')
assert_equal(ps.pedigree.sample_id.tolist(), pf.pedigree.sample_id.tolist(), 'Incorrect pedigree sample ID array')
assert_equal(ps.pedigree, pf.pedigree, 'Incorrect pedigree')
assert_equal(ps.genotype, pf.genotype, 'Incorrect genotype')
assert_equal(ps.haplotype, pf.haplotype, 'Incorrect haplotype')
self.assertTrue(ps.info.family_info != pf.info.family_info, 'Incorrect info family')
assert_not_equal(ps.info, pf.info, 'Incorrect info')
assert_not_equal(ps, pf, 'pf should have a family_info entry, ps shouldn''t')
Created on September 7, 2012
@author: Oren Livne <*****@*****.**>
============================================================
"""
import numpy as np, matplotlib.pyplot as plt, util
from impute import impute_test_util as itu
from impute.plot import plots
from impute.data import io
if __name__ == "__main__":
"""
--------------------------------------------------
Main program
--------------------------------------------------
"""
# Load nuclear family data
p = io.read_npz(itu.FAMILY2003_STAGE3)
f = p.families(genotyped=False)[0]
old_printoptions = np.get_printoptions()
np.set_printoptions(precision=2)
children = np.array([x for x in f.children_list if p.is_genotyped(x)])
h = p.haplotype.data
plots.plot_hap_corr_matrix(h, children, (0, 0))
plt.savefig("hap00.png")
plots.plot_hap_corr_matrix(h, children, (0, 1))
plt.savefig("hap01.png")
util.set_printoptions(old_printoptions)
def test_draw_pedigree(self):
'''Test famplot pedigree drawing tool integration.'''
p = read_npz(itu.NBHRS1298_STAGE4)
self.__test_draw_pedigree(p)
def setUp(self):
'''Load test data and expected results.'''
unittest.TestCase.setUp(self)
self.problem = io.read_npz(itu.NBHRS1298_STAGE4)
# Reset seed for deterministic results
np.random.seed(0)
def test_select_family(self):
'''Another test of creating a sub-problem by selecting a family..'''
p = io.read_npz(itu.FAMILY13_STAGE3)
p.sub_problem_of_parents(0, 1)
def test_draw_pedigree(self):
'''Test famplot pedigree drawing tool integration.'''
p = read_npz(itu.NBHRS1298_STAGE4)
self.__test_draw_pedigree(p)
def test_save_load_subproblem(self):
'''Check saving and loading a sub-problem with selected samples.'''
p = io.read_npz(itu.FAMILY13_STAGE3)
pp = p.sub_problem(p.families()[0].member_list)
self.__test_save_and_load_problem(pp, 'npz')
self.__test_save_and_load_problem(pp, 'plink')
def test_save_load_problem_with_problem_info(self):
'''Check that saving and loading a problem from file preserves the original object.'''
self.__test_save_and_load_problem(io.read_npz(itu.FAMILY13_STAGE3), 'npz')
self.__test_save_and_load_problem(io.read_npz(itu.FAMILY13_STAGE3), 'plink')
def test_select_family(self):
'''Another test of creating a sub-problem by selecting a family..'''
p = io.read_npz(itu.FAMILY13_STAGE3)
p.sub_problem_of_parents(0, 1)
Created on September 7, 2012
@author: Oren Livne <*****@*****.**>
============================================================
'''
import numpy as np, matplotlib.pyplot as plt, util
from impute import impute_test_util as itu
from impute.plot import plots
from impute.data import io
if __name__ == '__main__':
'''
--------------------------------------------------
Main program
--------------------------------------------------
'''
# Load nuclear family data
p = io.read_npz(itu.FAMILY2003_STAGE3)
f = p.families(genotyped=False)[0]
old_printoptions = np.get_printoptions()
np.set_printoptions(precision=2)
children = np.array([x for x in f.children_list if p.is_genotyped(x)])
h = p.haplotype.data
plots.plot_hap_corr_matrix(h, children, (0, 0))
plt.savefig('hap00.png')
plots.plot_hap_corr_matrix(h, children, (0, 1))
plt.savefig('hap01.png')
util.set_printoptions(old_printoptions)
============================================================
'''
from impute import impute_test_util as itu
from impute.plot import plots
import impute.ibd.ibd_child as ic
from impute.data import constants, io
if __name__ == '__main__':
'''
--------------------------------------------------
Main program
--------------------------------------------------
'''
# Load nuclear family data
p = io.read_npz(itu.FAMILY13_STAGE2)
f = p.families()[0]
# Calculate recombinations
parent_type = constants.PATERNAL
template = 2
parent = f.parents[parent_type]
(_, _, info) = ic.child_recombinations(p, f, parent_type, template, remove_errors=True)
# Different ways to visualize the results
print info.recombination_snp
info.plot(template=True)
info.plot(template=False)
plots.plot_family_comparison(p, f, parent_type, template=template)
plots.plot_family_comparison(p, f, parent_type)
