long_read_analysis_2

# initialise some variables to count short, medium length and long LTR5_Hs alignments.
long = 0
mid = 0
short = 0
count = 0
zero = 0

# specify the input path, this will be a directory containing multiple outputs from repeatmasker for a given tool.
input_path = filtered_ervcaller_path

# loop through each file in the directory.
for filename in os.listdir(input_path):
    count = count + 1
    curr_len = 0                        
    file = open(input_path + filename)   # open the file and loop through it.
    for line in file:
        if len(line.split()) > 10:       # ensure that the line is not empty. 
            if line.split()[0] != 'bit' and line.split()[0] != 'score':    # check that the current line is not the title line.
                if line.split()[9] == 'LTR5_Hs':                           # check if the line contains the LTR5_Hs identifier.
                    if line.split()[8] == '+':                             # find the length of the LTR5_Hs, we need to differentiate between forward and reverse orientation.
                        len_insert = int(line.split()[12]) - int(line.split()[11])
                    elif line.split()[8] == 'C':
                        len_insert = int(line.split()[12]) - int(line.split()[13])
                    if len_insert > curr_len:     # if the current LTR5_Hs length is longer than any previous LTR5_Hs from this file, update curr_len with this new length.
                        curr_len = len_insert
    if curr_len > 850:   # curr_len = the longest LTR5_Hs alignment from this file, fit it into one of the three length categories (or zero if there were no LTR5_Hs). 
        long += 1
    if curr_len > 400:
        mid += 1
    if curr_len > 0:
        short += 1
    if curr_len == 0:
        zero += 1
print('LTR5_Hs results:', long/count, mid/count, short/count, zero/count, count)

# the following script is identical to the above except it looks for lines which have any kind of LTR or ERV element, as opposed to jsut LTR5_Hs. 
long = 0
mid = 0
short = 0
count = 0
zero = 0
for filename in os.listdir(input_path):
    count = count + 1
    curr_len = 0
    file = open(input_path + filename)
    for line in file:
        if len(line.split()) > 10:
            if line.split()[0] != 'bit' and line.split()[0] != 'score':
                if line.split()[10] == 'LTR/ERVK' or line.split()[9] == 'HERVK-int'\
                or 'LTR5' in line.split()[9] or 'ERV' in line.split()[10]:
                    if line.split()[8] == '+':
                        len_insert = int(line.split()[12]) - int(line.split()[11])
                    elif line.split()[8] == 'C':
                        len_insert = int(line.split()[12]) - int(line.split()[13])
                    if len_insert > curr_len:
                        curr_len = len_insert
    if curr_len > 850:
        long += 1
    if curr_len > 400:
        mid += 1
    if curr_len > 0:
        short += 1
    if curr_len == 0:
        zero += 1
            
print('genreal ERV results:', long/count, mid/count, short/count, zero/count, count)