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

#!/usr/bin/env python
"""
Reads the REMc "-finalTable.csv" output file and makes a series of subdatasets
that reflect the pedigree structure of the way clusters break up.
"""

import sys
import os
import string
import glob


def reading_single_file(file_path):
    """
    Reads a file and generates a list of gene names.
    """
    with open(file_path, 'r') as data:
        attribute_line = data.readline().strip()
        attributes = attribute_line.split(',')

        gene_list = []
        for dataline in data:
            dataline = dataline.strip()
            elements = dataline.split(',')
            gene_list.append(elements[1])

    return gene_list


def writing_cluster_orf_list(list, output_dir, real_cluster_ori_name):
    """
    Writes a list of ORF names into a file in hierarchical series.
    """
    outfile_path = os.path.join(output_dir, f"{real_cluster_ori_name}.txt")
    with open(outfile_path, 'w') as outfile:
        for orf in list:
            outfile.write(orf.strip())
            outfile.write('\n')


def writing_cluster_results(attributes, orf_list, dic, output_dir, real_cluster_ori_name):
    """
    Writes clusters information into a series of files.
    """
    outfile_path = os.path.join(output_dir, f"{real_cluster_ori_name}-finaltable.csv")
    with open(outfile_path, 'w') as outfile:
        outfile.write(attributes)
        outfile.write('\n')
        for orf in orf_list:
            outfile.write(dic[orf.strip()].strip())
            outfile.write('\n')


def writing_ext_final_table(attributes, dic, ori_name_col_num, output_dir, output_file_name):
    """
    Writes the cluster name extensive final table.
    """
    outfile_path = os.path.join(output_dir, f"{output_file_name}-oriExtFinalTable.csv")
    with open(outfile_path, 'w') as outfile:
        outfile.write(attributes)
        outfile.write('\n')
        for orf in dic:
            elements = dic[orf].split(',')
            ori_name_list = elements[int(ori_name_col_num) - 1].strip().split(';')
            for ori_name in ori_name_list:
                elements.append(ori_name.strip())
            outfile.write(','.join(elements))
            outfile.write('\n')


def main():
    """
    Main function to parse the REMc -finalTable.csv output file.
    """
    try:
        data_file_path = sys.argv[1]
        output_path = sys.argv[2]
    except IndexError:
        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)

    try:
        with open(data_file_path, 'r') as data:
            attribute_line = data.readline().strip()
            attributes = attribute_line.split(',')
            orig_col = len(attributes) - 1

            data_dict = {}
            for dataline in data:
                dataline = dataline.strip()
                elements = dataline.split(',')
                data_dict[str.upper(elements[1].strip())] = ','.join(elements).upper()

    except FileNotFoundError:
        print('Input file does not exist')
        sys.exit(1)

    file_dict = {}
    for orf in data_dict:
        line = data_dict[orf].split(',')
        cluster_origin = line[int(orig_col) - 1].strip()
        cluster_identifier = cluster_origin.split(';')[0:-1]

        for i, identifier in enumerate(cluster_identifier):
            identifier = identifier.strip()
            if identifier not in file_dict:
                file_dict[identifier] = line[1]
            else:
                file_dict[identifier] = f"{file_dict[identifier]},{line[1]}"

    input_file_identifier = os.path.basename(data_file_path).split('.csv')[0]
    output_dir = os.path.join(output_path, input_file_identifier)
    os.makedirs(output_dir, exist_ok=True)

    # Writing the extensive ori name finaltable
    writing_ext_final_table(attribute_line, data_dict, orig_col, output_dir, input_file_identifier)

    # Writing the genelist files
    for cluster_name in file_dict:
        writing_cluster_orf_list(file_dict[cluster_name].split(','), output_dir, cluster_name)

        # Writing the cluster result files
        writing_cluster_results(attribute_line, file_dict[cluster_name].split(','), data_dict,
                                output_dir, cluster_name)


if __name__ == '__main__':
    main()