class ped_writer(unittest.TestCase):
def setUp(self):
self.kb = KB(driver='omero')('localhost', 'root', 'romeo')
def tearDown(self):
pass
def test_base(self):
def extract_data_sample(group, mset, dsample_name):
by_individual = {}
for i in self.kb.get_individuals(group):
gds = filter(lambda x: x.snpMarkersSet == mset,
self.kb.get_data_samples(i, dsample_name))
assert (len(gds) == 1)
by_individual[i.id] = gds[0]
return by_individual
study = self.kb.get_study('TEST01')
family = self.kb.get_individuals(study)
mset = self.kb.get_snp_markers_set(label='FakeTaqSet01')
gds_by_individual = extract_data_sample(study, mset,
'GenotypeDataSample')
pw = PedWriter(mset, base_path="./foo")
pw.write_map()
pw.write_family(study.id, family, gds_by_individual)
pw.close()
class ped_writer(unittest.TestCase):
def setUp(self):
self.kb = KB(driver="omero")("localhost", "root", "romeo")
def tearDown(self):
pass
def test_base(self):
def extract_data_sample(group, mset, dsample_name):
by_individual = {}
for i in self.kb.get_individuals(group):
gds = filter(lambda x: x.snpMarkersSet == mset, self.kb.get_data_samples(i, dsample_name))
assert len(gds) == 1
by_individual[i.id] = gds[0]
return by_individual
study = self.kb.get_study("TEST01")
family = self.kb.get_individuals(study)
mset = self.kb.get_snp_markers_set(label="FakeTaqSet01")
gds_by_individual = extract_data_sample(study, mset, "GenotypeDataSample")
pw = PedWriter(mset, base_path="./foo")
pw.write_map()
pw.write_family(study.id, family, gds_by_individual)
pw.close()
class App(object):
def __init__(self, host, user, passwd):
self.kb = KB(driver="omero")(host, user, passwd)
self.logger = logging.getLogger()
def compute(self, maker, model, release):
mset = self.kb.get_snp_markers_set(maker, model, release)
if not mset:
raise ValueError("SNPMarkersSet[%s,%s,%s] has not been defined." % (maker, model, release))
# projector = (np.arange(0, 100), np.array([101, 109]), np.arange(110,N))
# selector = kb.build_selector(
# s = self.kb.get_gdo_iterator(mset, selector, projector)
s = self.kb.get_gdo_iterator(mset)
# --
start = time.clock()
counts = algo.count_homozygotes(s)
print "counts on %d:" % counts[0], time.clock() - start
start = time.clock()
mafs = algo.maf(None, counts)
print "mafs on %d:" % counts[0], time.clock() - start
start = time.clock()
hwe = algo.hwe(None, counts)
print "hwe on %d:" % counts[0], time.clock() - start
class App(object):
def __init__(self, host, user, passwd):
self.kb = KB(driver='omero')(host, user, passwd)
self.logger = logging.getLogger()
def compute(self, maker, model, release):
mset = self.kb.get_snp_markers_set(maker, model, release)
if not mset:
raise ValueError('SNPMarkersSet[%s,%s,%s] has not been defined.' %
(maker, model, release))
# projector = (np.arange(0, 100), np.array([101, 109]), np.arange(110,N))
# selector = kb.build_selector(
# s = self.kb.get_gdo_iterator(mset, selector, projector)
s = self.kb.get_gdo_iterator(mset)
#--
start = time.clock()
counts = algo.count_homozygotes(s)
print 'counts on %d:' % counts[0], time.clock() - start
start = time.clock()
mafs = algo.maf(None, counts)
print 'mafs on %d:' % counts[0], time.clock() - start
start = time.clock()
hwe = algo.hwe(None, counts)
print 'hwe on %d:' % counts[0], time.clock() - start
class App(object):
def __init__(self, host, user, passwd,
study_label,
maker, model, release):
self.kb = KB(driver='omero')(host, user, passwd)
self.mset = self.kb.get_snp_markers_set(maker, model, release)
self.logger = logging.getLogger()
if not self.mset:
raise ValueError('SNPMarkersSet[%s,%s,%s] has not been defined.'
% (maker, model, release))
#--
alabel = 'load_genotypes-setup-%s' % time.time()
self.asetup = self.kb.factory.create(self.kb.ActionSetup,
{'label' : alabel,
'conf' : ''}).save()
#--
dmaker, dmodel, drelease = 'CRS4', 'load_genotypes', '0.1'
dlabel = '%s-%s-%s' % (dmaker, dmodel, drelease)
device = self.kb.get_device(dlabel)
if not device:
device = self.kb.factory.create(self.kb.Device,
{'label' : dlabel,
'maker' : dmaker,
'model' : dmodel,
'release' : drelease}).save()
self.device = device
#-- FIXME this will break if study is not defined.
self.study = self.kb.get_study(study_label)
def check_snp_markers_set(self, marker_types, marker_names):
self.logger.info('start checking snp_markers_set')
mdefs, msetc = self.kb.get_snp_markers_set_content(self.mset)
rs_labels = mdefs['rs_label']
for t, n in it.izip(marker_types, marker_names):
if t == 'M':
if not n in rs_labels:
msg = 'marker %s is not in the specified SNPMarkersSet' % n
self.logger.critical(msg)
raise ValueError(msg)
self.logger.info('done checking snp_markers_set')
def create_action(self, target):
conf = {'setup' : self.asetup,
'device' : self.device,
'actionCategory' : self.kb.ActionCategory.MEASUREMENT,
'operator' : 'Alfred E. Neumann',
'context' : self.study,
'target' : target,
}
action = self.kb.factory.create(self.kb.ActionOnVessel, conf).save()
return action
def create_data_sample(self, action, label):
conf = {'snpMarkersSet' : self.mset,
'label' : label,
'status' : self.kb.DataSampleStatus.USABLE,
'action' : action}
return self.kb.factory.create(self.kb.GenotypeDataSample, conf).save()
def load(self, pedfile, datfile, conf_value=1.0):
pr = PedReader(pedfile, datfile, conf_value)
self.check_snp_markers_set(pr.marker_types, pr.marker_names)
#--
self.logger.info('start loading from pedfile %s' % pedfile.name)
for x in pr:
sample = self.kb.get_vessel(x['sample_label'])
if not sample:
self.logger.error('No sample with label %s in VL' % x['sample_label'])
continue
action = self.create_action(sample)
avid = action.id
action.unload()
data_sample = self.create_data_sample(action, x['label'])
data_object = self.kb.add_gdo_data_object(avid, data_sample,
x['probs'], x['confs'])
self.logger.info('-- loaded %s' % x['label'])
self.logger.info('done loading from pedfile %s' % pedfile.name)
OME_USER = os.getenv('OME_USER', 'test')
OME_PASSWD = os.getenv('OME_PASSWD', 'test')
kb = KB(driver="omero")(OME_HOST, OME_USER, OME_PASSWD)
""" ..
The first thing we will do is to select a markers set. See FIXME:XXX
for its definition. We will first obtain an handle to it, and then
invoke a '.load_markers()' that will bring in memory the actual definition
data.
"""
mset_name = 'FakeTaqSet01'
mset0 = kb.get_snp_markers_set(label=mset_name)
mset0.load_markers()
""" ..
For the time being, we can think the SNPMarkerSet mset0 as analogous to an array
of markers. The following is a list of expressions that are expected
to be legal.
"""
len(mset0)
mset0[0::10]
mset0[11]
mset0[1].label
mset0[1].rs_label
OME_HOST = os.getenv('OME_HOST', 'localhost')
OME_USER = os.getenv('OME_USER', 'test')
OME_PASSWD = os.getenv('OME_PASSWD', 'test')
kb = KB(driver="omero")(OME_HOST, OME_USER, OME_PASSWD)
""" ..
The first thing we will do is to select a markers set. See FIXME:XXX
for its definition. We will first obtain an handle to it, and then
invoke a '.load_markers()' that will bring in memory the actual definition
data.
"""
mset_name = 'FakeTaqSet01'
mset0 = kb.get_snp_markers_set(label=mset_name)
mset0.load_markers()
""" ..
For the time being, we can think the SNPMarkerSet mset0 as analogous to an array
of markers. The following is a list of expressions that are expected
to be legal.
"""
len(mset0)
mset0[0::10]
mset0[11]
mset0[1].label
mset0[1].rs_label
""" ..
if args.logfile:
kwargs['filename'] = args.logfile
logging.basicConfig(**kwargs)
logger = logging.getLogger()
try:
host = args.host or vlu.ome_host()
user = args.user or vlu.ome_user()
passwd = args.passwd or vlu.ome_passwd()
except ValueError, ve:
logger.critical(ve)
sys.exit(ve)
kb = KB(driver="omero")(host, user, passwd)
logger.info("getting data samples")
ms = kb.get_snp_markers_set(label=args.marker_set)
if ms is None:
msg = "marker set %s not present in kb, bailing out" % args.marker_set
logger.critical(msg)
sys.exit(msg)
query = "from GenotypeDataSample g where g.snpMarkersSet.id = :id"
params = {"id": ms.omero_id}
gds = kb.find_all_by_query(query, params)
logger.info("found %d data samples for marker set %s" %
(len(gds), args.marker_set))
logger.info("updating dep tree")
kb.update_dependency_tree()
individuals = [get_individual(kb, ds) for ds in gds]
ds_by_ind_id = dict((i.id, ds) for i, ds in zip(individuals, gds))
logger.info("getting families")
families = get_all_families(kb)
class App(object):
def __init__(self, host, user, passwd, study_label, maker, model, release):
self.kb = KB(driver='omero')(host, user, passwd)
self.mset = self.kb.get_snp_markers_set(maker, model, release)
self.logger = logging.getLogger()
if not self.mset:
raise ValueError('SNPMarkersSet[%s,%s,%s] has not been defined.' %
(maker, model, release))
#--
alabel = 'load_genotypes-setup-%s' % time.time()
self.asetup = self.kb.factory.create(self.kb.ActionSetup, {
'label': alabel,
'conf': ''
}).save()
#--
dmaker, dmodel, drelease = 'CRS4', 'load_genotypes', '0.1'
dlabel = '%s-%s-%s' % (dmaker, dmodel, drelease)
device = self.kb.get_device(dlabel)
if not device:
device = self.kb.factory.create(
self.kb.Device, {
'label': dlabel,
'maker': dmaker,
'model': dmodel,
'release': drelease
}).save()
self.device = device
#-- FIXME this will break if study is not defined.
self.study = self.kb.get_study(study_label)
def check_snp_markers_set(self, marker_types, marker_names):
self.logger.info('start checking snp_markers_set')
mdefs, msetc = self.kb.get_snp_markers_set_content(self.mset)
rs_labels = mdefs['rs_label']
for t, n in it.izip(marker_types, marker_names):
if t == 'M':
if not n in rs_labels:
msg = 'marker %s is not in the specified SNPMarkersSet' % n
self.logger.critical(msg)
raise ValueError(msg)
self.logger.info('done checking snp_markers_set')
def create_action(self, target):
conf = {
'setup': self.asetup,
'device': self.device,
'actionCategory': self.kb.ActionCategory.MEASUREMENT,
'operator': 'Alfred E. Neumann',
'context': self.study,
'target': target,
}
action = self.kb.factory.create(self.kb.ActionOnVessel, conf).save()
return action
def create_data_sample(self, action, label):
conf = {
'snpMarkersSet': self.mset,
'label': label,
'status': self.kb.DataSampleStatus.USABLE,
'action': action
}
return self.kb.factory.create(self.kb.GenotypeDataSample, conf).save()
def load(self, pedfile, datfile, conf_value=1.0):
pr = PedReader(pedfile, datfile, conf_value)
self.check_snp_markers_set(pr.marker_types, pr.marker_names)
#--
self.logger.info('start loading from pedfile %s' % pedfile.name)
for x in pr:
sample = self.kb.get_vessel(x['sample_label'])
if not sample:
self.logger.error('No sample with label %s in VL' %
x['sample_label'])
continue
action = self.create_action(sample)
avid = action.id
action.unload()
data_sample = self.create_data_sample(action, x['label'])
data_object = self.kb.add_gdo_data_object(avid, data_sample,
x['probs'], x['confs'])
self.logger.info('-- loaded %s' % x['label'])
self.logger.info('done loading from pedfile %s' % pedfile.name)