def test_roh_mhmm_100pct(self):
# values correspond to start/stop/length/is_marginal
roh_expected = np.array([[1, 100, 100, True]], dtype=object)
fraction_expected = 1.0
gv = np.zeros((4, 2), dtype=np.int16)
pos = [1, 10, 50, 100]
roh, fraction = allel.roh_mhmm(gv, pos, contig_size=100)
aeq(roh.values, roh_expected)
assert fraction == fraction_expected
def test_roh_mhmm_0pct(self):
fraction_expected = 0.0
gv = np.zeros((4, 2), dtype=np.int16)
gv[2, 0] = 1
pos = [1, 10, 50, 100]
roh, fraction = allel.roh_mhmm(gv, pos, contig_size=100)
assert roh.shape[0] == 0
assert fraction == fraction_expected
def main():
if len(sys.argv) != 3:
print('vcf-file-name output-file')
sys.exit(1)
vcfFileName = sys.argv[1]
outputFile = sys.argv[2]
logger.debug('reading ' + vcfFileName)
vcf = allel.read_vcf(vcfFileName,
fields=[
'samples', 'calldata/GT', 'variants/ALT',
'variants/CHROM', 'variants/FILTER_PASS',
'variants/ID', 'variants/POS', 'variants/QUAL',
'variants/REF', 'variants/INFO'
])
logger.debug('reading data into genotype array')
genoArray = allel.GenotypeArray(vcf['calldata/GT'])
'''logger.debug('counting allele frequencies')
alleleFrequency = genoArray.count_alleles().to_frequencies()
logger.debug('calculating expected heterozygosity')
expectedHeterozygosity = allel.heterozygosity_expected(alleleFrequency, ploidy=2)
#print('expected: ' + str(expectedHeterozygosity))
logger.debug('calculating observed heterozygosity')
observedHeterozygosity = allel.heterozygosity_observed(genoArray)
#print('observed: ' + str(observedHeterozygosity))
logger.debug('calculating inbreeding coefficient')
inbreedingCoefficient = allel.inbreeding_coefficient(genoArray)
#print('ibc: ' + str(inbreedingCoefficient))
logger.debug('calculating delta heterozygosity')
diff = list()
for i in range(len(expectedHeterozygosity)):
diff.append(expectedHeterozygosity[i] - observedHeterozygosity[i])'''
logger.debug('finding runs of homozygosity')
numVariants = len(genoArray)
numSamples = len(vcf['samples'])
posArray = np.asarray([i for i in range(numVariants)])
isAccessible = np.asarray([True for i in range(numVariants)])
runsOfHomozygosity = dict()
for i in range(numSamples):
genoVector = genoArray[:, i]
roh = allel.roh_mhmm(genoVector, posArray, is_accessible=isAccessible)
print('roh = ' + str(roh))
if not roh[0].empty:
runsOfHomozygosity[i] = dict()
runsOfHomozygosity[i]['confidence'] = float(roh[1])
for j in range(len(roh[0])):
runsOfHomozygosity[i][j] = dict()
runsOfHomozygosity[i][j]['start'] = int(
roh[0].iloc[j]['start'])
runsOfHomozygosity[i][j]['stop'] = int(roh[0].iloc[j]['stop'])
runsOfHomozygosity[i][j]['is_marginal'] = bool(
roh[0].iloc[j]['is_marginal'])
'''logger.debug('saving ibc.txt')
np.savetxt(outputDir + '/ibc.txt', inbreedingCoefficient)
logger.debug('saving zygosityDelta.txt')
with open(outputDir + '/zygosityDelta.txt', 'w') as f:
for item in diff:
f.write("%s\n" % item)
f.close()'''
logger.debug('writing to ' + outputFile)
#roh = json.dumps(runsOfHomozygosity)
with open(outputFile, 'w') as f:
json.dump(runsOfHomozygosity, f)
f.close()