def mc(self):
r_obj = r['mc']
colnames = r.colnames(r_obj)
obj = array(r_obj).T
print(colnames)
return Struct(dict(zip(colnames, obj)))
def model(self):
def _fix_nones_list(column_list):
return [
None if isinstance(x, NARealType) else x for x in column_list
]
r_obj = r['model']
colnames = r.colnames(r_obj)
table = [_fix_nones_list(x) for x in r_obj]
return Struct(dict(zip(colnames, table)))
def parstat(self):
obj = array(r['ParStat']).T
colnames = r.colnames(obj)
properties = [
"xopt", "x1per", "x99per", "x10per", "x90per", "xminus", "xplus",
"mode", "mean", "sd", "skewness", "kurtosis"
]
parameter_values = [
dict(zip(properties, obj[i])) for i in range(len(colnames))
]
return Struct(dict(zip(colnames, parameter_values)))
def run(self, data_object):
"""Read canned dataset from R to a pandas dataframe
Returns:
data_object (DataObject): DataObject instance
terminate (bool): should we terminate the DAG? true or false
"""
dataset = self.node_config["dataset"]
logging.info("Reading {} from R".format(dataset))
try:
from rpy2.robjects.packages import importr, data
except ImportError: # pragma: no cover
raise ImportError(
"This example needs Rpy2."
"Please refer to the R requirements in the README"
)
datasets = importr("datasets")
r_env = data(datasets).fetch(dataset)
import rpy2.robjects as robjects
# why we do this:
# > data(euro)
# > euro
# ATS BEF DEM ESP FIM FRF IEP ITL LUF NLG PTE
# 13.760300 40.339900 1.955830 166.386000 5.945730 6.559570 0.787564 1936.270000 40.339900 2.203710 200.482000
#
# > as.data.frame(euro)
# euro
# ATS 13.760300
# BEF 40.339900
# DEM 1.955830
data = robjects.r("as.data.frame(%s)" % dataset)
# at time of writing, rpy2's R dataframe to pandas dataframe was not fully supported
# However, as python list() seems to work for FloatVector, StrVector, and FactorVector, let's use it
from rpy2.robjects import r
colnames = r.colnames(data)
pandas_data = {}
# convert each column of the R dataframe in turn
for i, colname in enumerate(colnames):
pandas_data[colname] = list(data[i])
# Unfortunately, some datasets have rownames that should be an ID column (e.g., see mtcars where rownames=names of the cars).
# This is the best we can do: pull it out as an additional column for each and every dataset
pandas_data["row_names"] = list(data.rownames)
df = pd.DataFrame(pandas_data)
data_object.add(self, df)
terminate = df.empty
return data_object, terminate
def _parse_assayData(assayData, assay):
"""Parse Rpy2 assayData (Environment object)
assayData: Rpy2 Environment object.
assay: An assay name indicating the data to be loaded.
Return a parsed expression dataframe (Pandas).
"""
pandas2ri.activate()
mat = assayData[assay] # rpy2 expression matrix object
data = pandas2ri.ri2py(mat)
features = pandas2ri.ri2py(r.rownames(mat))
samples = pandas2ri.ri2py(r.colnames(mat))
return pd.DataFrame(data, index=features, columns=samples)
def parstat(self):
r_obj = r['ParStat']
obj = array(r_obj).T
colnames = r.colnames(r_obj)
properties = [
"xopt", "x1per", "x99per", "x10per", "x90per", "xminus", "xplus",
"mode", "mean", "sd", "skewness", "kurtosis"
]
values = [dict(zip(properties, obj[i])) for i in range(len(colnames))]
matrix = [
[""] + properties,
]
for i, param in enumerate(colnames):
temp = [
param,
]
for prop in properties:
temp.append(values[i][prop])
matrix.append(temp)
return matrix
# Extract required arguments.
data = pd.read_table(snakemake.input.data, index_col=0) # Input Gene-by-Sample raw count data.
condition = pd.read_table(snakemake.input.condition, index_col=0, names=['condition']) # Input condition file which indicates to which condition each sample belongs.
logger.info('%d(genes) x %d(samples) data matrix and %d sample conditions are given.' % (data.shape[0], data.shape[1], len(condition.index)))
logger.debug('Headers: %s...' % ' '.join(data.columns[:3]))
logger.debug('Gene identifiers: %s...' % ' '.join(data.index[:3]))
intersecting_samples = [sample for sample in data.columns if sample in condition.index]
data = data[intersecting_samples]
condition = list(condition.loc[intersecting_samples].condition.values)
logger.info('%d samples will be used for DEG discovery.' % len(intersecting_samples))
r_data_matrix = r['data.matrix'](pandas2ri.py2ri(data))
r_samples = r.colnames(r_data_matrix)
r_conditions = ro.FactorVector(condition)
logger.debug('Computing size factors.')
r_size_factors = ebseq.MedianNorm(r_data_matrix)
logger.info('Discovering DEGs.')
logger.info('Running EBTest.')
num_iteration = 0
while True:
# Increase iteration numbers if the conditons are not met.
# Hopefully most of the tie, 10 iterations will be enough for convergence.
num_iteration += 10
r_eb_out = ebseq.EBTest(Data=r_data_matrix, Conditions=r_conditions, sizeFactors=r_size_factors, maxround=num_iteration)
logger.info('Running GetDEResults. (FDR cutoff = %.3f)' % cutoff)