combine-motevo-results.py

##############################################################################
#   AUTHOR: Krish Agarwal
#   AFFILIATION: University_of_Basel
#   CONTACT: akrish136@gmail.com
#   CREATED: 14-07-2020
#   LICENSE: Apache_2.0
##############################################################################

##### Importing necessary libraries #####
import os
import sys
import pandas as pd
import csv
from argparse import ArgumentParser, RawTextHelpFormatter

##### Using argparse to get input from command line #####
parser = ArgumentParser(description=__doc__, formatter_class=RawTextHelpFormatter)

parser.add_argument(
    "--input_directories",
    nargs="+",
    dest="input_directories",
    help="input directory for the script to search",
    required=True,
    metavar="DIR",
)

parser.add_argument(
    "--filename",
    dest="filename",
    help="filename to be searched in input_directories",
    required=True,
    metavar="NAME",
)

parser.add_argument(
    "--outfile",
    dest="outfile",
    help="location and name of the tsv file",
    required=True,
    metavar="FILE",
)

try:
    options = parser.parse_args()
except Exception:
    parser.print_help()

if len(sys.argv) == 1:
    parser.print_help()
    sys.exit(1)

##### Initializing dictionaries and list #####
tabb = dict()
tabb["strand"] = []
tabb["binding_position"] = []
tabb["binding_posterior"] = []
tabb["pwm_id"] = []
tabb["binding_region"] = []
tabb["binding_sequence"] = []
tabb["LogLik_ratio_fg_bg"] = []
tabb["fasta_record"] = []

##### Taking input from commandline #####
filenameparam = str(options.filename)
dirrs = options.input_directories
outdir = str(options.outfile)

##### Sorting the input directories so as to prevent different orders for same input #####
dirrs = sorted(dirrs)

rel_outfile = os.path.relpath(outdir)

##### Processing each directory one by one using a for loop #####
for dirr in dirrs:
    abs_path = os.path.join(dirr, filenameparam)  # absolute path of the input file

    filename = os.path.relpath(abs_path)  # relative path of input file

    i = 0  # this variable will be used to check whether we are at the first line of the record or second

    file = open(filename, "r")  # open the file in read mode

    for each in file:  # processing the file line by line

        each_word = each.split(" ")  # splitting the line into individual words

        if i % 2 == 0:  # if i is even then we are at the first line

            ##### Appending values #####
            tabb["binding_position"] = tabb["binding_position"] + [each_word[0]]
            tabb["strand"] = tabb["strand"] + [each_word[1]]
            tabb["binding_posterior"] = tabb["binding_posterior"] + [each_word[2]]
            tabb["pwm_id"] = tabb["pwm_id"] + [each_word[3]]
            tabb["binding_region"] = tabb["binding_region"] + [each_word[4]]

        else:  # if i is odd then we are at the second line

            ##### Appending values #####
            tabb["binding_sequence"] = tabb["binding_sequence"] + [each_word[0]]
            tabb["LogLik_ratio_fg_bg"] = tabb["LogLik_ratio_fg_bg"] + [each_word[1]]
            tabb["fasta_record"] = tabb["fasta_record"] + [each_word[2]]

        i = i + 1  # increment the value of i after we are done processing each line

list_items = [
    "pwm_id",
    "binding_position",
    "binding_sequence",
    "binding_posterior",
    "LogLik_ratio_fg_bg",
]  # these will be the headers in the final tsv file

tempDict1 = (
    {}
)  # initializing empty dictionary that will contain data only of the headers mentioned in list_items

for item in list_items:

    tempDict = {item: pd.Series(tabb[item])}  # appending list of values for each key
    tempDict1.update(tempDict)  # concating dictionary

df = pd.DataFrame(tempDict1)  # convert dictionary to dataframe
df = df.sort_values(by=['binding_posterior'], ascending=False) # sort values according to binding posterior in descending

df.to_csv(rel_outfile, index=False, sep="\t", header=list_items)  # creating a tsv file