def features_selection():
"""
-> Perform a features selection procedure:
- get arguments from the command line
- i : the input file (mandatory)
- o : the output folder (mandatory)
- r : the name of the report file (optionnal, default is report.html)
- run analysis (analysis type set to "all")
- write report
"""
## Collect arguments
parser = argparse.ArgumentParser(prog='netabio_stuff') # not sure what does the instanciation
parser.add_argument('-i', nargs='?', help='the input data file')
parser.add_argument('-o', nargs='?', help='the output directory')
parser.add_argument('-r', nargs='?', help='the name of the report file')
args = parser.parse_args()
## Init optionnal arguments
if(not args.r):
args.r = "report.html"
collected_inputs = {'i': args.i,
'r': args.r,
'o': args.o}
## Get the name of the prepared data file
report_file = str(collected_inputs["o"])+"/"+str(collected_inputs["r"])
data_file = str(collected_inputs["i"])
result_folder = str(collected_inputs["o"])
## Run analysis
feature_selection.run_analyser(data_file, result_folder, "all")
## write report
feature_selection.write_html_report(report_file, data_file, result_folder)
def features_selection_workspace():
"""
[IN PROGRESS]
-> Grand bazar, scavange stuff from this function
-> TODO:
- get a list of args
- implement the feature selection procedure ...
- include the fudction in entry point
"""
## Collect arguments
parser = argparse.ArgumentParser(prog='netabio_stuff') # not sure what does the instanciation
parser.add_argument('-i', nargs='?', help='the input data file')
parser.add_argument('-gp', nargs='?', help='the group position (label) in the reformated file')
parser.add_argument('-gn', nargs='?', help='the group name (colname of the label) in the reformated file')
parser.add_argument('-ph', nargs='?', help='the phase we want in the reformated dile')
parser.add_argument('-o', nargs='?', help='the output directory')
args = parser.parse_args()
## Init optionnal arguments
if(not args.gp):
args.gp = "start"
if(not args.gn):
args.gn = "Disease"
if(not args.ph):
args.ph = "all"
collected_inputs = {'i': args.i,
'gp': args.gp,
'gn': args.gn,
'ph': args.ph,
'o': args.o}
## Prepare data file
#data_extraction.generate_Luminex_data_file(collected_inputs['i'], group_pos=collected_inputs['gp'], group_name=collected_inputs['gn'], phase=collected_inputs['ph'])
## Get the name of the prepared data file
report_file = "C:\\Users\\NaturalKiller01\\Desktop\\Nathan\\Spellcraft\\TRASH\\report.html"
data_file = "C:\\Users\\NaturalKiller01\\Desktop\\Nathan\\Spellcraft\\Luminex_data_NA_filtered.csv"
result_folder = "C:\\Users\\NaturalKiller01\\Desktop\\Nathan\\Spellcraft\\TRASH"
## Run analysis
feature_selection.run_analyser(data_file, result_folder, "all")
## write report
feature_selection.write_html_report(report_file, data_file, result_folder)
def run():
## general parameters
data_file_name = TEST_DATA
separator = ","
test_value = "53,7%"
input_vector = ["5", "3", "4", "6", "89", "5", "6"]
col_number = 1
header_detected = True
output_folder = "output"
analysis = "all"
report_file = "report.html"
## biotoolbox module
print "[TEST] Testing biotoolbox module from netabio package"
biotoolbox.detect_file_format(data_file_name)
biotoolbox.change_file_format(data_file_name, separator)
biotoolbox.fix_file_name(data_file_name)
## na_manager module
print "[TEST] Testing na_manager module from netabio package"
na_manager.check_NA_proportion_in_file(data_file_name)
na_manager.display_NA_proportions(data_file_name)
na_manager.filter_NA_values(data_file_name)
## quality_control module
print "[TEST] Testing quality_control module from netabio package"
quality_control.check_pourcentages(test_value)
quality_control.looking_for_outliers(input_vector, col_number)
quality_control.basic_check(data_file_name)
quality_control.check_standard_deviation(data_file_name)
quality_control.check_zscore(data_file_name, separator, header_detected)
## feature_selection module
print "[TEST] Testing feature_selection module from netabio package"
feature_selection.run_analyser(data_file_name, output_folder, analysis)
feature_selection.write_html_report(report_file, data_file_name,
output_folder)