if a != '':
normal_spin_axis = a
else:
assert False, "%s option is not supported" % opt
if input_path == None:
print 'No input file was provided.\nYou need to specify an input file\n(e.g. backSPIN -i path/to/your/file/foo.cef)\n'
sys.exit()
if outfiles_path == None:
print 'No output file was provided.\nYou need to specify an output file\n(e.g. backSPIN -o path/to/your/file/bar.cef)\n'
sys.exit()
try:
if verbose:
print 'Loading file.'
input_cef = CEF_obj()
input_cef.readCEF(input_path)
data = array(input_cef.matrix)
if feature_fit:
if verbose:
print "Performing feature selection"
ix_features = feature_selection(data, feature_genes, verbose=verbose)
if verbose:
print "Selected %i genes" % len(ix_features)
data = data[ix_features, :]
input_cef.matrix = data.tolist()
input_cef.row_attr_values = atleast_2d( array( input_cef.row_attr_values ))[:,ix_features].tolist()
input_cef.update()
from Cef_tools import CEF_obj import numpy as numpy input_CEF = "BackSPIN_Output_CEF" output_Clusters = "CLUSTERS_OUTFILE" output_Gene_Sets = "GENE_SET_OUTFILE" cef = CEF_obj() cef.readCEF(input_CEF) your_array = cef.matrix attribute_values_list2 = cef.col_attr_names attribute_values_list1 = cef.row_attr_values rowvals_list = numpy.transpose(attribute_values_list1) varListr = ['CellId', 'bRUNID'] numLevel = numpy.alen(attribute_values_list2) covals_last = numpy.transpose(cef.col_attr_values[numLevel - 1]) covals_gene = numpy.transpose(cef.col_attr_values[0]) covals = numpy.column_stack((covals_gene, covals_last)) covalsFinal = numpy.vstack((varListr, covals)) numpy.savetxt(output_Clusters, covalsFinal, delimiter=",", fmt="%s") numpy.savetxt(output_Gene_Sets, rowvals_list, delimiter=",", fmt="%s") # End writing clusters and gene sets.
outfiles_path = a
elif opt in ('-c', '--clus_attr'):
clustering_attribute = str(a)
elif opt in ('-p', '--processes'):
n_processes = int(a)
elif opt in ('-s', '--stanfolder'):
stan_folder = str(a)
else:
assert False, "%s option is not supported" % opt
# Determine the format of the input and read it
input_format = input_path.split('.')[-1]
if input_format == 'cef':
# Load the cef file
from Cef_tools import CEF_obj
cef = CEF_obj()
try:
print('Loading CEF')
cef.readCEF(input_path)
except:
print('Error in loading CEF file')
elif input_format == 'loom':
# Load the loom file
import loompy
print("Connecting to the .loom file")
ds = loompy.connect(input_path)
else:
print(
'The format specified is not accepted. Please provide a .loom or .cef file'
)
# in Zeisel et al. Cell types in the mouse cortex and hippocampus revealed by
# single-cell RNA-seq Science 2015 (PMID: 25700174, doi: 10.1126/science.aaa1934).
#
# Building using pyinstaller:
# pyinstaller -F backSPIN.py -n backspin-mac-64-bit
'''
from Cef_tools import CEF_obj
from numpy import genfromtxt
import numpy as numpy
import csv
input_file = "SAMPLE_FILE_FROM_SEQGEQ"
output_file = "BackSPIN_Input_CEF_fName"
date_ran = "DATE_TIME_SCRIPT_RAN"
Arr_values = genfromtxt(input_file, delimiter=',')
gene_names = genfromtxt(input_file, delimiter=',', dtype=str)
with open(input_file, 'rb') as f:
reader = csv.reader(f)
cell_names = next(reader)
cef = CEF_obj()
cef.add_header('Date ran', date_ran)
cef.set_matrix(Arr_values[1:, 1:])
cef.add_row_attr('Gene', gene_names[1:, :1])
cef.add_col_attr('CellName', cell_names[1:])
cef.writeCEF(output_file)