#!/usr/bin/python


"""
Second BioCreative challenge Protein Protein Interaction,
IAS subtask submission format check script.

"""


###########################
# IMPORT PYTHON LIBRARIES #
###########################
import string, re, os, random, sys
from optparse import OptionParser



####################
####################
def get_nr_pmidlist(red_pmid_list):

    nr_list = []

    for pmid in red_pmid_list:
        if pmid not in nr_list:
            nr_list.append(pmid)

    return nr_list



###########################
# INPUT ERROR CHECK
###########################
def capture_input_errors(pred_true,pred_false,pmid_test_file):
    """Check for missing input arguments """


    if pred_true == None:

        print "Please specify the PPI relevant prediction file!"
        sys.exit(2)

    if pred_false == None:
        
        print "Please specify the PPI not relevant prediction file!"
        sys.exit(2)

    if pmid_test_file == None:
        print "Please specify the PMID check file!"
        sys.exit(2)




########################
# GET WHOLE PMID LIST  # 
########################
def get_pmid_list(pmid_test_file):
    """GET THE LIST OF PUBMED IDS IN THE DATA COLLECTION """

    pmid_test_file = str(pmid_test_file)
    pmid_test_file = string.strip(pmid_test_file)

    
    try:
        pmid_filename = open(pmid_test_file,'r')
    except:
        print "Error could not open file: ", pmid_test_file 


    
    pmid_list = []

    while 1:

        file_line = pmid_filename.readline()

        if file_line:
            if len(file_line) > 0:
                pmid = str(file_line)
                pmid = string.strip(pmid)
                pmid_list.append(pmid)
                    
        else:
            break


    return pmid_list








###############################
# GET/CHECK PREDICTION FILES  # 
###############################
def get_pred_list(pred_file):
    """GET THE PREDICTION RUN FILES """

    pred_file = str(pred_file)
    pred_file = string.strip(pred_file)

    
    try:
        filename = open(pred_file,'r')
    except:
        print "Error could not open file: ", pmid_test_file 


                        
    team_list      = []
    run_id_list    = []
    task_id_list   = []
    pred_type_list = []
    pred_rank_list = []
    pred_pmid_list = []

    while 1:

        file_line = filename.readline()

        line_counter = 1
        if file_line:
            if len(file_line) > 0:
                lines = str(file_line)
                line_split = string.split(lines,'\t')
                
                if len(line_split) != 6:
                    print "CHECK SUBMISSION FORMAT OF LINE: ", line_counter
                    print lines
                    print line_split
                    break

                if len(line_split) == 6:
                    
                    team_id   = line_split[0]
                    run_id    = line_split[1]
                    task_id   = line_split[2]
                    pred_type = line_split[3]
                    pred_rank = line_split[4]
                    pred_pmid = line_split[5]

                    col_cnt = 1
                    for column in line_split:
                        column = str(column)
                        column = string.strip(column)
                        
                        if len(column) < 1:
                            print "MISSING VALUE FOR ", col_cnt
                            print "LINE NUMBER: ", line_counter
                            print lines
                            print line_split
                            break


                        col_cnt = col_cnt + 1

                        
                    team_id   = str(team_id)
                    team_id   = string.strip(team_id)
                    
                    run_id    = str(run_id)
                    run_id    = string.strip(run_id)
                    
                    task_id   = str(task_id)
                    task_id   = string.strip(task_id)
                    
                    pred_type = str(pred_type)
                    pred_type = string.strip(pred_type)
                    
                    pred_rank = str(pred_rank)
                    pred_rank = string.strip(pred_rank)
                    
                    pred_pmid = str(pred_pmid)
                    pred_pmid = string.strip(pred_pmid)
                    

                    team_list.append(team_id)
                    run_id_list.append(run_id)
                    task_id_list.append(task_id)
                    pred_type_list.append(pred_type)
                    pred_rank_list.append(pred_rank)
                    pred_pmid_list.append(pred_pmid)
                    

            line_counter = line_counter + 1
            
                    
                
                    
        else:
            break


        
    return team_list, run_id_list, task_id_list, pred_type_list, pred_rank_list, pred_pmid_list







###########################
# MAIN FUNCTION OF SCRIPT #
###########################
if __name__ == '__main__':


  
    
    # Define script input options
    parser = OptionParser()
    
    parser.add_option("--t",default=None,help="Predicted as PPI relevant",dest="pred_true")
    parser.add_option("--f",default=None,help="Predicted as not PPI relevant",dest="pred_false")
    parser.add_option("--i",default=None,help="PMID check file",dest="pmid_test_file")

    
    # Read the options provided by the user
    (options, args) = parser.parse_args(sys.argv[1:])


    
    
    if sys.argv[1:] == []:
        print "Please specify input options!",
        print '\n'
        print "Example input for team 60, run 1:"
        print "./formatcheck_bc2_ppi_ias_V01.py --t BC2_PPI_IAS_T60_BC2_PPI_1_T --f BC2_PPI_IAS_T60_BC2_PPI_1_F --i bc2_ppi_ias_pmid_test.txt"
        print '\n',
        print "Where:"
        print "--t refers to the entries predicted as PPI relevant"
        print "--f refers to the entries predicted as not PPI relevant"
        print "--i refers to the test set PMID check file, i.e. bc2_ppi_ias_pmid_test.txt"
        print '\n'
        
  
        
        sys.exit(0)

        
    pred_true      = options.pred_true
    pred_false     = options.pred_false
    pmid_test_file = options.pmid_test_file

    #check input argument 
    capture_input_errors(pred_true,pred_false,pmid_test_file)

    #get pmid list
    pmid_list = get_pmid_list(pmid_test_file)

    #get PPI relevant predicted
    team_t_list, run_id_t_list, task_id_t_list, pred_type_t_list, pred_rank_t_list, pred_pmid_t_list = get_pred_list(pred_true)

  
    #get PPI not relevant predicted
    team_f_list, run_id_f_list, task_id_f_list, pred_type_f_list, pred_rank_f_list, pred_pmid_f_list = get_pred_list(pred_false)


    ###################
    #CHECK TEAM NAMES #
    ###################
    nr_t_names = get_nr_pmidlist(team_t_list)
    nr_f_names = get_nr_pmidlist(team_f_list)

    if len(nr_t_names) > 1:
        print "CHECK TEAM NAMES COLUMN IN ", pred_true
        print nr_t_names
        
    if len(nr_f_names) > 1:
        print "CHECK TEAM NAMES COLUMN IN ", pred_false
        print nr_f_names
        
    
    
    #######################
    #CHECK PREDICTION RUN #
    #######################
    nr_t_run = get_nr_pmidlist(run_id_t_list)
    nr_f_run = get_nr_pmidlist(run_id_f_list)

    if len(nr_t_run) > 1:
        print "CHECK RUN ID COLUMN IN ", pred_true
        print nr_t_run
        
    if len(nr_f_run) > 1:
        print "CHECK RUN ID COLUMN IN ", pred_false
        print nr_f_run
    
    
    
  
    
    #######################
    #CHECK TASK ID COLUMN #
    #######################
    nr_t_task = get_nr_pmidlist(task_id_t_list)
    nr_f_task = get_nr_pmidlist(task_id_f_list)

    if len(nr_t_task) > 1:
        print "CHECK TASK ID COLUMN IN ", pred_true
        print nr_t_task
    
    if len(nr_f_task) > 1:
        print "CHECK TASK ID COLUMN IN ", pred_false
        print nr_f_task
        
    
        
    #####################
    #CHECK RANK  COLUMN #
    #####################
    nr_t_rank = get_nr_pmidlist(pred_rank_t_list)
    nr_f_rank = get_nr_pmidlist(pred_rank_f_list)

    if len(nr_t_rank) != len(pred_rank_t_list):
        print "CHECK RANK COLUMN IN ", pred_true
        print nr_t_rank
        
    if len(nr_f_rank) != len(pred_rank_f_list):
        print "CHECK RANK COLUMN IN ", pred_false
        print nr_f_rank
        
        
    #######################
    #CHECK CORRECT  PMIDS #
    #######################
    for t_pmid in pred_pmid_t_list:
        if t_pmid not in pmid_list:
            print "RELEVANT PREDICTED PMID ERROR , NOT IN PMID CHECK FILE: ", t_pmid
        
        
    for f_pmid in pred_pmid_f_list:
        if f_pmid not in pmid_list:
            print "RELEVANT PREDICTED PMID ERROR , NOT IN PMID CHECK FILE: ", f_pmid
            
        
    
    #####################################
    # CHECK PMID PREDICTIONS REDUNDANCY #
    #####################################
    nr_t_pmid = get_nr_pmidlist(pred_pmid_t_list)
    nr_f_pmid = get_nr_pmidlist(pred_pmid_f_list)

    if len(nr_t_pmid) != len(pred_pmid_t_list):
        print "CHECK PMID REDUNDANCY IN ", pred_true
        print nr_t_pmid
        
    if len(nr_f_pmid) != len(pred_pmid_f_list):
        print "CHECK PMID REDUNDANCY IN ", pred_false
        print nr_f_pmid
        

        
    ###################################
    # CHECK MULTIPLE PMID PREDICTIONS #
    ###################################
    for pmid_true in pred_pmid_t_list:
        if pmid_true in pred_pmid_f_list:
            print "PMID PREDICTION ERROR, A GIVEN PMID CAN NOT BE IN BOTH TRUE AND FALSE SET: ", pmid_true
            

            
    ##########################
    # CHECK PREDICTION START #
    ##########################
    if pred_rank_t_list[0] != '1':
        print "RANK IN PPI RELEVANT PREDICTIONS STARTS WITH 1, NOT WITH: ", pred_rank_t_list[0]
        
    if pred_rank_f_list[0] != '1':
        print "RANK IN PPI NOT RELEVANT PREDICTIONS STARTS WITH 1, NOT WITH: ", pred_rank_f_list[0]
        

    #######################
    # CHECK MISSING PMIDS #
    #######################
    all_pred = pred_pmid_t_list + pred_pmid_f_list
    if len(all_pred) != len(pmid_list):
        print "CHECK FOR MISSING PMID PREDICTIONS, YOU HAVE TO SUBMIT PREDICTIOSN FOR ALL ENTRIES!"
    

    total_t = len(pred_pmid_t_list)
    total_f = len(pred_pmid_f_list)
    total_entries = len(pmid_list)
    
    ############################
    # PRINT PREDICTION SUMMARY #
    ############################
    print '\n'
    print '======================================================================='
    print '                   PREDICTION SUMMARY OF ', nr_t_names[0]
    print '======================================================================='
    print "FILE CONTAINING THE PPI RELEVANT PREDICTIONS: "
    print pred_true
    print "FILE CONTAINING THE PPI NOT RELEVANT PREDICTIONS: "
    print pred_false
    print '-----------------------------------------------------------------------'
    print "TOTAL NUMBER OF ENTRIES IN DATA SET:               ", len(pmid_list)
    print "TOTAL NUMBER OF ENTRIES PREDICTED AS PPI RELEVANT: ", len(pred_pmid_t_list)
    print "TOTAL NUMBER OF ENTRIES PREDICTED AS PPI RELEVANT: ", len(pred_pmid_f_list)
    print "PERCENTAGE OF TOTAL SET PREDICTED AS PPI RELEVANT  ", float(total_t) / float(total_entries)
    print "PERCENTAGE OF TOTAL SET PREDICTED AS NOT PPI RELEVANT  ", float(total_f) / float(total_entries)
    print '======================================================================='