def _remember_dict(r, robj=None):
new = ro.ListVector({name:Remember._remember_item(x) for name, x in r.items()})
if robj is None:
return new
else:
return ro.ListVector({**dict(robj.items()), **dict(new.items())})
def testNewListVector(self):
vec = robjects.ListVector({'a': 1, 'b': 2})
self._testNewListVector(vec)
s = (('a', 1), ('b', 2))
vec = robjects.ListVector(s)
self._testNewListVector(vec)
it = iter(s)
vec = robjects.ListVector(s)
self._testNewListVector(vec)
def viper(self, data_mx, subunit_set, subunit_tfms):
from rpy2 import robjects
from rpy2.robjects import r, pandas2ri
from rpy2.robjects.conversion import localconverter
from rpy2.robjects.packages import importr
base = importr('base')
try:
vp = importr("viper")
except:
base.source("http://www.bioconductor.org/biocLite.R")
biocinstaller = importr("BiocInstaller")
biocinstaller.biocLite("viper")
vp = importr("viper")
# Conduct VIPER analysis
r_networks = robjects.ListVector.from_length(len(subunit_tfms))
for i, subunit_tfm in enumerate(subunit_tfms):
regulons = []
for subunit in subunit_set:
tfmode = robjects.FloatVector(
np.asarray(subunit_tfm[subunit]).astype(float))
tfmode.names = robjects.StrVector(subunit_set[subunit])
likelihood = robjects.FloatVector(
np.repeat(1.0, len(subunit_set[subunit])))
regulon = robjects.ListVector({
'tfmode': tfmode,
'likelihood': likelihood
})
regulons.append(regulon)
# Generate R regulon
r_networks[i] = robjects.ListVector(
zip(subunit_set.keys(), regulons))
# Generate R matrix
mx_nr, mx_nc = data_mx.shape
mx_vec = robjects.FloatVector(data_mx.values.transpose().reshape(
(data_mx.size)))
r_mx = robjects.r.matrix(mx_vec, nrow=mx_nr, ncol=mx_nc)
r_mx.rownames = robjects.StrVector(data_mx.index)
r_mx.colnames = robjects.StrVector(data_mx.columns)
# Compute VIPER profile
vpres = vp.viper(r_mx,
r_networks,
verbose=False,
minsize=1,
cores=self.threads)
with localconverter(robjects.default_converter + pandas2ri.converter):
vpres_df = robjects.conversion.rpy2py(vpres)
pd_mx = pd.DataFrame(vpres_df, columns=vpres.colnames)
pd_mx['query_id'] = vpres.rownames
return (pd_mx)
def call_function(id, function, add, *args, **kwargs):
ro.globalenv["args"] = ro.ListVector([["wazzup", x] for x in args])
ro.globalenv["kwargs"] = ro.ListVector([[name, x] for name, x in kwargs.items()])
ro.r("names(args) <- NULL")
if add:
fun = f"{id} = do.call(dynwrap::{function}, c(list({id}), args, kwargs));NULL"
ro.r(fun)
else:
fun = f"{id} = do.call(dynwrap::{function}, c(args, kwargs))"
ro.r(fun)
return id
def sarima_test(steps, path):
index_name, my_trend = parse_csv(path)
dta = pd.DataFrame(my_trend)
dta.index = index_name
dta = dta.rename(columns={0: 'search'})
r_df = com.convert_to_r_dataframe(dta)
y = stats.ts(r_df)
order = R.IntVector((1, 1, 1))
season = R.ListVector({'order': R.IntVector((0, 1, 0)), 'period': 52})
model = stats.arima(y[-5 * 52:-steps], order=order, seasonal=season)
f = forecast.forecast(model, h=steps)
future = [var for var in f[3]]
y_pred = np.array(future)
y_true = np.array(my_trend[-steps:])
metrics_result = {
'sarima_MAE': metrics.mean_absolute_error(y_true, y_pred),
'sarima_MSE': metrics.mean_squared_error(y_true, y_pred),
'sarima_MAPE': np.mean(np.abs((y_true - y_pred) / y_true)) * 100
}
p1 = plt.plot(my_trend[-steps:], '*-')
p2 = plt.plot(future)
# p1 = plt.plot(index_name,my_trend,'r-')
# p2 = plt.plot(index_name_future,future,'g-')
plt.ylabel('Search Intensity')
plt.xlabel('Year')
plt.title('Search Prediction of ' + path.split('/')[-1][:-4])
plt.legend((p1[0], p2[0]), ["Actual", "Predicted"], loc=1)
plt.grid(True)
# print metrics_result['sarima_MAPE']
return metrics_result['sarima_MAPE']
def get_gcindices_r(countsGeneLabels,
correctBackground=False,
remove_empty=True):
"""
>>> indices_r = get_gcindices(counts.index,correctBackground=False) # doctest: +SKIP
TODO overlapping with genesets2indices_r => refactor code
"""
limma = importr('limma')
gc = LSD.get_msigdb6()
if correctBackground:
gc = {
gsc: {
gs: [g for g in gc[gsc][gs] if g in countsGeneLabels]
for gs in gc[gsc]
}
for gsc in gc
}
countsGeneLabels_r = ro.StrVector(countsGeneLabels)
gc_indices_r = {
gsc: limma.ids2indices(ro.ListVector(gc[gsc]),
countsGeneLabels_r,
remove_empty=remove_empty)
for gsc in gc
}
return gc_indices_r
def identify(self, polarity, in_file, databases, non_empty, mzmatch_params,
mzmatch_outputs):
group_dict = {}
for group_label, index, description, files, abspath in non_empty:
group_dict[group_label] = robjects.StrVector(files)
groups = robjects.ListVector(group_dict)
# turns out that 'stds.xml.db' always has a value, e.g. /home/pimp/media/projects/10/analysis_38/stds_db.xml
# regardless of whether the file exists or not.
# this behaviour is different from the old pipeline? So here we set it to R NULL if the file doesn't exist
stds_xml_file = self.get_value(mzmatch_outputs, 'stds.xml.db')[0]
stds_xml_file = os.path.abspath(stds_xml_file)
if not os.path.isfile(stds_xml_file):
logger.info('%s is not found, setting stds.xml.db to NULL',
stds_xml_file)
self.set_value(mzmatch_outputs, 'stds.xml.db', robjects.r("NULL"))
args = {
'in_file': in_file,
'databases': databases,
'groups': groups,
'mzmatch.outputs': mzmatch_outputs,
'mzmatch.params': mzmatch_params,
'polarity': polarity
}
raw_data = run_r('Pimp.identify.metabolites', **args)
return raw_data
def setREnv(self, call, tsname='r_timeseries', inline=False, **kwargs):
command = '{}({}'.format(call, tsname)
for key, item in kwargs.items():
if isinstance(item, bool):
if inline: rinline = "TRUE" if item else "FALSE"
else:
ro.globalenv[
key] = ro.rinterface.TRUE if item else ro.rinterface.FALSE
elif isinstance(item, list) and all(
isinstance(x, float) for x in item):
if inline: rinline = item # This is not yet correct!
else: ro.globalenv[key] = pandas2ri.FloatSexpVector(item)
elif isinstance(item, list) and all(
isinstance(x, int) for x in item):
rinline = None
ro.globalenv[key] = pandas2ri.IntSexpVector(item)
elif isinstance(item, dict):
rinline = None
ro.globalenv[key] = ro.ListVector(item)
else:
try:
if inline is False: ro.globalenv[key] = item
else:
if isinstance(item, str): rinline = '\"' + item + '\"'
else: rinline = item
except:
logging.error('Variable {} - Traceback - {}'.format(
key, self.rtracebackerror()))
if inline: command += ", {}={}".format(key, rinline)
else: command += ", {}={}".format(key, key)
command += ")"
return command
def get_deseq_result(self, contrast=None, **kwargs):
'''
DESeq2: result(dds, contrast)
making a dds.deseq_result pandas dataframe
'''
if contrast:
if len(contrast) == 3:
R_contrast = robjects.vectors.StrVector(np.array(contrast))
else:
if len(contrast) != 2:
raise ValueError('Contrast must be length of 3 or 2')
R_contrast = robjects.ListVector(
{None: con
for con in contrast})
logger.info('Using contrast: %s' % contrast)
self.result = deseq.results(self.dds,
contrast=R_contrast,
**kwargs) # Robject
else:
self.result = deseq.results(self.dds, **kwargs) # R object
self.deseq_result = to_dataframe(self.result) # R dataframe
with localconverter(robjects.default_converter + pandas2ri.converter):
self.deseq_result = robjects.conversion.rpy2py(
self.deseq_result) ## back to pandas dataframe
self.deseq_result[self.gene_column] = self.gene_id.values
def eval(self, x):
x = np.array(x, ndmin=1)
if "X" in self.log10:
x_new = ro.ListVector({"x": ro.FloatVector(np.log10(x))})
else:
x_new = ro.ListVector({"x": ro.FloatVector(x)})
y_pred = self.mgcv.predict_gam(self.spline, newdata=x_new)
if "Y" in self.log10:
y_pred = 10 ** np.array(y_pred)
else:
y_pred = np.array(y_pred)
#
if len(y_pred) == 1:
return y_pred[0]
else:
return y_pred
def testReprNonVectorInList(self):
vec = robjects.ListVector(
OrderedDict((
('a', 1),
('b', robjects.Formula('y ~ x')),
)))
s = repr(vec).split('\n')
self.assertEqual('[IntVector, Formula]', s[2].strip())
def sarima(steps, path):
index_name, my_trend = parse_csv(path)
dta = pd.DataFrame(my_trend)
dta.index = index_name
dta = dta.rename(columns={0: 'search'})
#dta.plot(figsize=(10,4))
#==============================================================================
# check stationarity
#==============================================================================
#r_df = com.convert_to_r_dataframe(DataFrame(dta))
#y = stats.ts(r_df)
#ad = tseries.adf_test(y, alternative="stationary", k=52)
#a = ad.names[:5]
#{ad.names[i]:ad[i][0] for i in xrange(len(a))}
#==============================================================================
# check the seasonality
#==============================================================================
#diff1lev = dta.diff(periods=1).dropna()
#diff1lev.plot(figsize=(12,6))
#diff1lev_season = diff1lev.diff(52).dropna()
#r_df = com.convert_to_r_dataframe(DataFrame(diff1lev_season))
#diff1lev_season1lev = diff1lev_season.diff().dropna()
#==============================================================================
# check stationarity after difference
#==============================================================================
#y = stats.ts(r_df)
#ad = tseries.adf_test(y, alternative="stationary", k=52)
#a = ad.names[:5]
#{ad.names[i]:ad[i][0] for i in xrange(len(a))}
#==============================================================================
# plot acf and pacf
#==============================================================================
#fig = plt.figure(figsize=(12,8))
#ax1 = fig.add_subplot(211)
#fig = sm.graphics.tsa.plot_acf(diff1lev_season1lev.values.squeeze(), lags=150, ax=ax1)
#ax2 = fig.add_subplot(212)
#fig = sm.graphics.tsa.plot_pacf(diff1lev_season1lev, lags=150, ax=ax2)
#fig
r_df = com.convert_to_r_dataframe(dta)
y = stats.ts(r_df)
order = R.IntVector((1, 1, 1))
season = R.ListVector({'order': R.IntVector((0, 1, 0)), 'period': 52})
a = time.time()
model = stats.arima(y, order=order, seasonal=season)
print time.time() - a
f = forecast.forecast(model, h=steps)
future = [var for var in f[3]]
dt = date_range(dta.index[-1], periods=len(future) + 1,
freq='W')[1:] #создаем индекс из дат
pr = Series(future, index=dt)
# dta.plot(figsize=(12,6))
# pr.plot(color = 'red')
return index_name, dt, my_trend, future
def _run_gsea(df, genesets, method='ssgsea', verbose=False, **kwargs):
rdata = r('as.matrix')(df)
rgenesets = robjects.ListVector(genesets)
res = r('gsva')(rdata, rgenesets, method=method, verbose=verbose, **kwargs)
py_res = pandas2ri.ri2py_dataframe(res)
py_res.index = r('rownames')(res)
# py_res.columns = r('colnames')(res)
py_res.columns = df.columns
return py_res
def getKruskal(wt_rankpt_dist, mut_rankpt_dist, mut_wt_conn_dist):
return robjects.r["kruskal.test"](robjects.ListVector({
'a':
robjects.FloatVector(wt_rankpt_dist),
'b':
robjects.FloatVector(mut_rankpt_dist),
'c':
robjects.FloatVector(mut_wt_conn_dist)
}))[2][0]
def test_repr_nonvectorinlist():
vec = robjects.ListVector(
OrderedDict((
('a', 1),
('b', robjects.Formula('y ~ x')),
)))
s = repr(vec)
assert s.startswith("R object with classes: (\'RTYPES.VECSXP',) "
"mapped to:\n[IntVector, Formula]")
def test_repr_nonvectorinlist():
vec = robjects.ListVector(
OrderedDict((
('a', 1),
('b', robjects.Formula('y ~ x')),
)))
s = repr(vec)
assert s.startswith("R object with classes: ('list',) mapped to:%s"
"[IntSexpVector, LangSexpVector]" % os.linesep)
def predict_unstructured(self, data, **kwargs):
def _r_is_character(r_val):
_is_character = ro.r("is.character")
return bool(_is_character(r_val))
def _r_is_raw(r_val):
_is_raw = ro.r("is.raw")
return bool(_is_raw(r_val))
def _r_is_null(r_val):
return r_val == ro.rinterface.NULL
def _cast_r_to_py(r_val):
# TODO: consider checking type against rpy2 proxy object like: isinstance(list_data_kwargs, ro.vectors.ListVector)
# instead of calling R interpreter
if _r_is_null(r_val):
return None
elif _r_is_raw(r_val):
return bytes(r_val)
elif _r_is_character(r_val):
# Any scalar value is returned from R as one element vector,
# so get this value.
return str(r_val[0])
else:
raise DrumCommonException(
"Can not convert R value {} type {}".format(
r_val, type(r_val)))
def _rlist_to_dict(rlist):
if _r_is_null(rlist):
return None
return {str(k): _cast_r_to_py(v) for k, v in rlist.items()}
data_binary_or_text = data
if UnstructuredDtoKeys.QUERY in kwargs:
kwargs[UnstructuredDtoKeys.QUERY] = ro.ListVector(
kwargs[UnstructuredDtoKeys.QUERY])
# if data_binary_or_text is str it will be auto converted into R character type;
# otherwise if it is bytes, manually convert it into byte vector (raw)
r_data_binary_or_text = data_binary_or_text
if isinstance(data_binary_or_text, bytes):
r_data_binary_or_text = ro.vectors.ByteVector(data_binary_or_text)
kwargs_filtered = {k: v for k, v in kwargs.items() if v is not None}
list_data_kwargs = r_handler.predict_unstructured(
model=self._model, data=r_data_binary_or_text, **kwargs_filtered)
if isinstance(list_data_kwargs, ro.vectors.ListVector):
ret = _cast_r_to_py(list_data_kwargs[0]), _rlist_to_dict(
list_data_kwargs[1])
else:
raise DrumCommonException(
"Wrong type returned in unstructured mode: {}".format(
type(list_data_kwargs)))
return ret
def test_repr_nonvectorinlist():
vec = robjects.ListVector(
OrderedDict((
('a', 1),
('b', robjects.Formula('y ~ x')),
)))
s = repr(vec).split(os.linesep)
assert s[1].startswith("R classes: ('list',)")
assert s[2].startswith("[IntSexpVector, LangSexpVector]")
def toR(something):
if isinstance(something, list):
if isinstance(something, float):
return ri.FloatSexpVector(something)
elif isinstance(something, int):
return ri.IntSexpVector(something)
else:
return ri.StrSexpVector(something)
elif isinstance(something, dict):
return ro.ListVector(something)
return something
def heatmap_annotation_key(self, name, colors):
'''
generates data frame for color key for the annotation
from a dict
'''
keyColors = ro.StrVector([c for c in colors.values()])
keyColors.names = colors.keys()
key = OrderedDict()
key[name] = keyColors
return ro.ListVector(key)
def convert_dict(obj):
if all([isinstance(x, str) for x in obj]):
return ro.StrVector(obj)
elif all([isinstance(x, int) | isinstance(x, float) for x in obj]):
return ro.IntVector(obj)
elif all([isinstance(x, bool) for x in obj]):
return ro.BoolVector(obj)
elif all([isinstance(x, float) for x in obj]):
return ro.FloatVector(obj)
return ro.ListVector(obj)
def on_pbExecute_clicked(self):
R.r(''' source('Rasterise_dev_61.R') ''')
for i in range(0,self.ui.tableWidget.rowCount()):
self.Drivername[i] = self.ui.tableWidget.cellWidget(i,0).text()
self.DictList[str(self.Drivername[i])] = self.__filename[i]
self.DictLen = len(self.DictList)
drvs = R.ListVector(self.DictList)
genSummary = R.r['genrateStatisticalSummary']
self.res = genSummary(str(self.DriverType),self.__T0File,self.__T1File,drvs,int(self.ui.leNAValue.text()))
self.ui.teSummary.setPlainText(str(self.res))
def d_bse_(d, N, type="quant"):
v = robjects.FloatVector(numpy.array(d["se"])**2)
f = robjects.FloatVector(d["frequency"].values)
b = robjects.FloatVector(d["beta"].values)
s = _s(d, N)
return robjects.ListVector({
"beta": b,
"varbeta": v,
"MAF": f,
"N": s,
"type": type
})
def train_elastic_net_wrapper(features_data_, features_, d_, data_annotation_, x_w=None, prune=True, nested_folds=10):
x = numpy.array([features_data_[v] for v in features_.id.values])
dimnames = robjects.ListVector(
[(1, robjects.StrVector(d_["individual"])), (2, robjects.StrVector(features_.id.values))])
x = robjects.r["matrix"](robjects.FloatVector(x.flatten()), ncol=features_.shape[0], dimnames=dimnames)
y = robjects.FloatVector(d_[data_annotation_.gene_id])
nested_folds = robjects.FloatVector([nested_folds])
#py2ri chokes on None.
if x_w is None:
res = train_elastic_net(y, x, n_train_test_folds=nested_folds)
else:
res = train_elastic_net(y, x, penalty_factor=x_w, n_train_test_folds=nested_folds) # observation weights, not explanatory variable weight :( , x_weight = x_w)
return pandas2ri.ri2py(res[0]), pandas2ri.ri2py(res[1])
def _filter_and_values_to_RList(d):
"""`d` is a dictionary of filters: values. Returns a StrVector and
a ListVector of StrVectors"""
# Could use ListVector directly with the dict, but want to guarantee
# positional order of filters and values
f = robjects.StrVector(list(d.keys()))
v = robjects.ListVector(
rpy2.rlike.container.TaggedList(
d.values(),
tags=list(d.keys())
)
)
return f, v
def pynlsfit(valuelis, formulastr='', startvalues=list(), filename='nlsfit.txt', gformat='pdf'):
"""
nonlinear fit of function to data
"""
rconsole = rpystatinit()
rnonlinfit = rconsole("rnonlinfit")
dataframe = pyobj2dataframe(valuelis)
from rpy2.robjects.packages import importr
grdevices = importr('grDevices')
graphplotfile = fhutils.renamefilename(filename, suffix=gformat)
grdevices.pdf(file=graphplotfile)
nlfit = rnonlinfit(data=dataframe, formulastr=formulastr, startvalues=robjects.ListVector(dict(startvalues)))
grdevices.dev_off()
nlfit = rlisttodic(nlfit)
return nlfit
def get_deseq_result(self, contrast=None, **kwargs):
self.comparison = deseq.resultsNames(self.dds)
if contrast:
if len(contrast) == 3:
contrast = robjects.numpy2ri.numpy2ri(np.array(contrast))
else:
assert len(contrast) == 2, 'Contrast must be length of 3 or 2'
contrast = robjects.ListVector({None: con for con in contrast})
print('Using contrast: ', contrast)
self.deseq_result = deseq.results(self.dds,
contrast=contrast,
**kwargs)
else:
self.deseq_result = deseq.results(self.dds, **kwargs)
self.deseq_result = to_dataframe(self.deseq_result)
self.deseq_result = pandas2ri.ri2py_dataframe(
self.deseq_result) ## back to pandas dataframe
self.deseq_result[self.gene_column] = self.gene_id.values
def get_deseq_result(self, contrast=None, **kwargs):
self.comparison = deseq.resultsNames(self.dds)
if contrast:
if len(contrast) == 3:
contrast = robjects.numpy2ri.numpy2ri(np.array(contrast))
else:
assert len(contrast) == 2, 'Contrast must be length of 3 or 2'
contrast = robjects.ListVector({None: con for con in contrast})
print('Using contrast: ', contrast)
self.deseq_result = deseq.results(self.dds,
contrast=contrast,
**kwargs)
else:
self.deseq_result = deseq.results(self.dds, **kwargs)
self.deseq_result = to_dataframe(self.deseq_result)
self.deseq_result = conversion.rpy2py(self.deseq_result)
return (self.deseq_result)
def get_predictions(self, train_period_begin, train_period_end, prediction_count):
#print 'Getting training data from', train_period_begin, 'to', train_period_end
self.training_data = self.full_data[train_period_begin: train_period_end]
self.forecast_period = prediction_count
#print self.dataset.iloc[train_period_begin:train_period_end]
r_df = com.convert_to_r_dataframe(DataFrame(self.dataset.iloc[train_period_begin:train_period_end]))
y = PricePredictor.stats.ts(r_df)
orderR = R.IntVector((1,0,1))
season = R.ListVector({'order': R.IntVector((1,0,1)), 'period' : 24})
model = PricePredictor.stats.arima(y, order = orderR, seasonal=season,method="ML")
f = PricePredictor.forecast.forecast(model, h=self.forecast_period)
#print f
#print "\n"
predValues = []
for item in f.items():
if item[0] == 'mean':
for value in item[1].items():
predValues.append(value[1])
return predValues
def genesets2indices_r(genesets, geneLabels, remove_empty=True):
"""
genesets should be a dictionary of geneset lists, and geneLabels a list of gene labels
>>> from .tests.test_retro import testGenesets, testCountsGenelabels
>>> print(genesets2indices_r(testGenesets,testCountsGenelabels))
[[1]]
[1] 1 2
<BLANKLINE>
[[2]]
[1] 2 3
<BLANKLINE>
<BLANKLINE>
TODO test function for gene level conversion between python and R
"""
limma = importr('limma')
genesets_r = ro.ListVector(
{gs: ro.StrVector(genesets[gs])
for gs in genesets})
return limma.ids2indices(genesets_r, geneLabels, remove_empty=remove_empty)
