class Experiment(object):
'''A validation experiment: start with a Problem object, clear a certain portion of the
data, run a phaser, and cross-check the hap results against the original genotype data.'''
#---------------------------------------------
# Constructors
#---------------------------------------------
def __init__(self, problem, fraction=None, test_index=None):
'''Initialize an experiment to be run on a problem, clearing out 'fraction' of the data. If test_index
is specified, these specific test indices are used; otherwise a random fraction is generated.
If test_index = 'hap', data is read from problem.h (haplotype array). The entire array
is considered as a test array, but nothing is zeroed out. Useful for phasing result stats.'''
# Create a working copy of the problem. Only the data is copied.
if not (fraction is not None) ^ (test_index is not None):
raise ValueError('Must specify fraction or test_index')
self.problem = Problem(problem.pedigree, problem.genotype.copy())
self.h = self.problem.h
# Create test set; save original genotypes in g_orig
if test_index is None:
self.fraction = fraction
self.g_orig, i = clear_random_portion(self.problem.genotype.data, fraction)
elif test_index == 'hap':
# Don't clear anything; call everything a test index.
h = problem.h
i = tuple(util.flattened_meshgrid(range(h.shape[0]), range(h.shape[1])))
self.g_orig = problem.g
self.h = h
self.fraction = 1.0
else:
self.g_orig, i = clear_index(self.problem.g, test_index)
self.fraction = (1.0 * i[0].size) / (self.h.shape[0] * self.h.shape[1])
self.num_tests = i[0].size
self.test_index = i
self.r_orig = recode.recode_single_genotype(self.g_orig)
self.fill = self.problem.fill_fraction()[:, SAMPLE]
self.__recode_single_genotype = None
#---------------------------------------------
# Methods
#---------------------------------------------
def __repr__(self):
return 'Experiment[%s, fraction=%.2f%%]' % (repr(self.problem), self.fraction)
def run(self, phaser, params=None):
'''Run phaser (or more generally, a handler) on a problem.'''
phaser.run(self.problem, params=params)
self.fill = self.problem.fill_fraction()[:, 1]
def num_test_genotypes(self, field):
'''Return the number of genotypes in which both alleles were called, broken by field (SNP=0, sample=1).'''
return self.__group_by_field(np.arange(len(self.test_index[0])), field)
def where_called(self):
'''Return the indices of genotypes in which both alleles were called.'''
# Positive entries of r = called entries
return recode.where_called(self.recoded_genotype)[0]
def called(self, field):
'''Return the number of genotypes in which both alleles were called, broken by field (SNP=0, sample=1).'''
return self.__group_by_field(self.where_called(), field)
#---------------------------------------------
# Properties
#---------------------------------------------
@property
def test_orig(self):
'''Return the original set of deleted test genotypes.'''
i = self.test_index
return recode.recode_single_genotype(self.h[i[SNP], i[SAMPLE], :])
@property
def test_called(self):
'''Return the called set of haplotypes corersponding to the test genotypes.'''
i = self.test_index
return self.h[i[SNP], i[SAMPLE], :]
@property
def recoded_genotype(self):
'''Return the genotype test set, recoded as a single number of allele pair.'''
if self.__recode_single_genotype is None:
self.__recode_single_genotype = recode.recode_single_genotype(self.test_called)
return self.__recode_single_genotype
@property
def total_called(self):
'''Return the number of genotypes in which both alleles were called.'''
return len(self.where_called())
@property
def total_partial_called(self):
'''Return the number of genotypes in which one alleles was called.'''
# Positive entries of r = called entries
return len(recode.where_partial_called(self.recoded_genotype)[0])
@property
def total_errors(self):
'''Return the number of genotypes that were called incorrectly. (A genotype is an allele pair.)'''
# Count entries that were called (positive AND are different than the corresponding original value
return len(recode.where_error(self.recoded_genotype, self.r_orig)[0])
@property
def total_partial_errors(self):
'''Return the number of genotypes that were called incorrectly. (A genotype is an allele pair.)'''
# Count entries that were called (positive AND are different than the corresponding original value
# This happens when hap=(0,1), genotype=(2,2) or hap(0,2), genotype=(1,1)
return len(recode.where_partial_error(self.recoded_genotype, self.r_orig)[0])
@property
def full_call_fraction(self):
'''Return the % of correctly fully-called test genotypes.'''
return (1.0 * self.total_called) / self.num_tests
@property
def partial_call_fraction(self):
'''Return the % of erroneously half-called test genotypes.'''
return (1.0 * self.total_partial_called) / self.num_tests
@property
def full_error_fraction(self):
'''Return the % of erroneously fully-called test genotypes.'''
return (1.0 * self.total_errors) / self.num_tests
@property
def partial_error_fraction(self):
'''Return the % of erroneously half-called test genotypes.'''
return (1.0 * self.total_partial_errors) / self.num_tests
@property
def stats(self):
'''Return a tuple containing all experiment statistics: fraction, all call rates, run time.'''
return (self.fraction,
self.full_call_fraction,
self.partial_call_fraction,
self.full_error_fraction,
self.partial_error_fraction)
#---------------------------------------------
# Private Methods
#---------------------------------------------
def __group_by_field(self, i, field):
'''Group a test index subset i by field (SNP=0, sample=1).'''
size = self.problem.genotype.data.shape[field]
group_count = util.dict_to_array(statutil.group_by_value(self.test_index[field][i]))
result = np.zeros((size,), dtype=int)
result[group_count['k']] = group_count['v']
return result
class Experiment(object):
'''A validation experiment: start with a Problem object, clear a certain portion of the
data, run a phaser, and cross-check the hap results against the original genotype data.'''
#---------------------------------------------
# Constructors
#---------------------------------------------
def __init__(self, problem, fraction=None, test_index=None):
'''Initialize an experiment to be run on a problem, clearing out 'fraction' of the data. If test_index
is specified, these specific test indices are used; otherwise a random fraction is generated.
If test_index = 'hap', data is read from problem.h (haplotype array). The entire array
is considered as a test array, but nothing is zeroed out. Useful for phasing result stats.'''
# Create a working copy of the problem. Only the data is copied.
if not (fraction is not None) ^ (test_index is not None):
raise ValueError('Must specify fraction or test_index')
self.problem = Problem(problem.pedigree, problem.genotype.copy())
self.h = self.problem.h
# Create test set; save original genotypes in g_orig
if test_index is None:
self.fraction = fraction
self.g_orig, i = clear_random_portion(self.problem.genotype.data,
fraction)
elif test_index == 'hap':
# Don't clear anything; call everything a test index.
h = problem.h
i = tuple(
util.flattened_meshgrid(range(h.shape[0]), range(h.shape[1])))
self.g_orig = problem.g
self.h = h
self.fraction = 1.0
else:
self.g_orig, i = clear_index(self.problem.g, test_index)
self.fraction = (1.0 * i[0].size) / (self.h.shape[0] *
self.h.shape[1])
self.num_tests = i[0].size
self.test_index = i
self.r_orig = recode.recode_single_genotype(self.g_orig)
self.fill = self.problem.fill_fraction()[:, SAMPLE]
self.__recode_single_genotype = None
#---------------------------------------------
# Methods
#---------------------------------------------
def __repr__(self):
return 'Experiment[%s, fraction=%.2f%%]' % (repr(
self.problem), self.fraction)
def run(self, phaser, params=None):
'''Run phaser (or more generally, a handler) on a problem.'''
phaser.run(self.problem, params=params)
self.fill = self.problem.fill_fraction()[:, 1]
def num_test_genotypes(self, field):
'''Return the number of genotypes in which both alleles were called, broken by field (SNP=0, sample=1).'''
return self.__group_by_field(np.arange(len(self.test_index[0])), field)
def where_called(self):
'''Return the indices of genotypes in which both alleles were called.'''
# Positive entries of r = called entries
return recode.where_called(self.recoded_genotype)[0]
def called(self, field):
'''Return the number of genotypes in which both alleles were called, broken by field (SNP=0, sample=1).'''
return self.__group_by_field(self.where_called(), field)
#---------------------------------------------
# Properties
#---------------------------------------------
@property
def test_orig(self):
'''Return the original set of deleted test genotypes.'''
i = self.test_index
return recode.recode_single_genotype(self.h[i[SNP], i[SAMPLE], :])
@property
def test_called(self):
'''Return the called set of haplotypes corersponding to the test genotypes.'''
i = self.test_index
return self.h[i[SNP], i[SAMPLE], :]
@property
def recoded_genotype(self):
'''Return the genotype test set, recoded as a single number of allele pair.'''
if self.__recode_single_genotype is None:
self.__recode_single_genotype = recode.recode_single_genotype(
self.test_called)
return self.__recode_single_genotype
@property
def total_called(self):
'''Return the number of genotypes in which both alleles were called.'''
return len(self.where_called())
@property
def total_partial_called(self):
'''Return the number of genotypes in which one alleles was called.'''
# Positive entries of r = called entries
return len(recode.where_partial_called(self.recoded_genotype)[0])
@property
def total_errors(self):
'''Return the number of genotypes that were called incorrectly. (A genotype is an allele pair.)'''
# Count entries that were called (positive AND are different than the corresponding original value
return len(recode.where_error(self.recoded_genotype, self.r_orig)[0])
@property
def total_partial_errors(self):
'''Return the number of genotypes that were called incorrectly. (A genotype is an allele pair.)'''
# Count entries that were called (positive AND are different than the corresponding original value
# This happens when hap=(0,1), genotype=(2,2) or hap(0,2), genotype=(1,1)
return len(
recode.where_partial_error(self.recoded_genotype, self.r_orig)[0])
@property
def full_call_fraction(self):
'''Return the % of correctly fully-called test genotypes.'''
return (1.0 * self.total_called) / self.num_tests
@property
def partial_call_fraction(self):
'''Return the % of erroneously half-called test genotypes.'''
return (1.0 * self.total_partial_called) / self.num_tests
@property
def full_error_fraction(self):
'''Return the % of erroneously fully-called test genotypes.'''
return (1.0 * self.total_errors) / self.num_tests
@property
def partial_error_fraction(self):
'''Return the % of erroneously half-called test genotypes.'''
return (1.0 * self.total_partial_errors) / self.num_tests
@property
def stats(self):
'''Return a tuple containing all experiment statistics: fraction, all call rates, run time.'''
return (self.fraction, self.full_call_fraction,
self.partial_call_fraction, self.full_error_fraction,
self.partial_error_fraction)
#---------------------------------------------
# Private Methods
#---------------------------------------------
def __group_by_field(self, i, field):
'''Group a test index subset i by field (SNP=0, sample=1).'''
size = self.problem.genotype.data.shape[field]
group_count = util.dict_to_array(
statutil.group_by_value(self.test_index[field][i]))
result = np.zeros((size, ), dtype=int)
result[group_count['k']] = group_count['v']
return result
