server-scripts/qhtcp-workflow/apps/python/DconJG2.py

#!/usr/bin/env python
"""
Improved code to determine the origin column (OligCol) without user input of the argument
and removed the former sys.argv[2]. JWR 22_0816
"""
"""
this code can be used for the parse of the REMc "-finalTable.csv" output file 
to make a series of subdatasets, which reflect the pedigree structure 
of the way cluters breaking up. 
"""

import sys, os, string, glob

try:
   data_file_Path = sys.argv[1]
   #print data_file_Path
   #cluster_ori_col_num = sys.argv[2]
   output_path = sys.argv[2]
except: 
   print ('Usage: python parse_clustering_result_to_Pedigree_Dataset_and_genelist.py /datasetPath/datasetfilename cluster_origin_column_num output_path_name')
   print ('Data file not found')
   sys.exit(1)

#define a function to reading files and generate the list
def read_file(file_path):
    with open(file_path, 'r') as file:
        attributes = file.readline().strip().split(',')
        gene_list = [elements[1] for line in file for elements in [line.strip().split(',')]]
    return gene_list

# define a function to write the list into a file named in hierarchical series
def write_cluster_orf_list(orf_list, output_dir, cluster_name):
    cluster_file_path = os.path.join(output_dir, f"cluster_name, 'txt')
    with open(cluster_file_path, 'w') as outfile:
        for orf in orf_list:
            outfile.write(orf.strip() + '\n')

# define a function to write the clusters information into a series of files
def write_cluster_results(attributes, orf_list, data_dict, output_directory, cluster_name):
    file_path = os.path.join(output_directory, f"{cluster_name}-finaltable.csv")
    with open(file_path, 'w') as output_file:
        output_file.write(attributes)
        output_file.write('\n')
        for orf in orf_list:
            output_file.write(data_dict[orf.strip()].strip())
            output_file.write('\n')

# define a function to write the cluster name origina extensive final table 
def write_extended_final_table(attributes, data, ori_name_column_number, output_directory, output_file_name):
    output_file_path = os.path.join(output_directory, f"{output_file_name}-oriExtFinalTable.csv")
    with open(output_file_path, 'w') as output_file:
        output_file.write(attributes)
        output_file.write('\n')
        for orf in data:
            elements = data[orf].split(',')
            ori_name_list = elements[int(ori_name_column_number)-1].strip().split(';')
            for ori_name in ori_name_list:
                elements.append(orii_name.strip())
            output_file.write(','.join(elements))
            output_file.write('\n')

# Read the data file
try:
   data = open(data_file_Path,'r')
except OSError:
   print ('input file does not exists')

 
# first the title line would be read and kept
attributes = data.readline().strip().split(',')
print(attributes)
print(len(attributes))
OrigCol= len(attributes) - 1
print(OrigCol)

# then the data
data_dict = {}
for data_line in data:
    data_line = data_line.strip()
    line_elements = data_line.split(',')
    orf_identifier = line_elements[1].strip().upper()
    data_dict[orf_identifier] = ','.join(line_elements).upper()
data.close()
#print dataDic
   
print ("OrigCol is ", str(OrigCol))
fileDic = {}
for orf in dataDic:
    line = dataDic[orf].split(',')
    #read the cluster name len(attributes)
    clusterOrigin = line[int(OrigCol) - 1]
    
    #clusterOrigin = line[int(cluster_ori_col_num) - 1]
    #print clusterOrigin
    clusterOrigin = clusterOrigin.strip()
    #print clusterOrigin
    clusterIdentifier = clusterOrigin.split(';')[0:-1]
    #print clusterIdentifier
    
    for identifier in clusterIdentifier:
        identifier = identifier.strip()
        upper_identifier = identifier.upper()
        if upper_identifier not in fileDic:
            fileDic[upper_identifier] = line[1]
        else:
            fileDic[upper_identifier] += ',' + line[1]
	    
input_file_identifier = data_file_Path.strip().split('/')[-1].strip().split('.csv')[-3]

#make the output folder
try:
    os.mkdir(str(output_path)+str(input_file_identifier))
except OSError:
	
	
print ('dir exists')

#Writing the extensive ori name finaltable
Writing_ext_final_table(attributeLine, dataDic,str(OrigCol),str(output_path)+str(input_file_identifier), str(input_file_identifier))
#Writing_ext_final_table(attributeLine, dataDic,cluster_ori_col_num,str(output_path)+str(input_file_identifier), str(input_file_identifier))

#write the genelist files
for cluster_name in fileDic:
    #print fileDic[cluster_name].split(',')
    Writing_clusterORF_list(fileDic[cluster_name].split(','), str(output_path)+str(input_file_identifier), cluster_name)                        
    #write the cluster result files
    Writing_cluster_results(attributeLine, fileDic[cluster_name].split(','), dataDic,str(output_path)+str(input_file_identifier),cluster_name)