Python: Parse binary patterns from file and get the unique binary patterns

Python: Parse binary patterns from file and get unique patterns.py

# assuming that the file is tab-delimited ('\t')
delim = '\t'

# binary file path:
binary_patterns_filename = "binary_patterns.txt"

# get all of the lines in the binary patterns file
lines = [line.rstrip() for line in open(binary_patterns_filename, "r")]

# 1st line should contain the headers
headers = lines[0].split(delim)

print "All headers:", headers
# 1st column should contain the name/index of pattern/isolate/etc
print "First header:", headers[0]
# starting at the 1st line after the header line
# get all of the unique patterns and store in a dictionary
unique_fings = {} # dictionary to store all unique fingerprints
for line in lines[1:]:
  # split the current line based on the delimiter (e.g. tabs)
	delim_split = line.split(delim)
	# get the element in the first column
    first_element = delim_split[0]
    # concatenate all of the '0's and '1's to get the fingerprint
    fing = "".join(delim_split[1:])
    # check if the fingerprint isn't already in the unique_fings dictionary
    if fing not in unique_fings:
    	# add the new fingerprint and start a list containing the first element
    	# string ID
        unique_fings[fing] = [first_element]
    else:
    	# if the fingerprint is already present in the unique_fings dictionary
    	# add the first element string ID to the fingerprint 
        unique_fings[fing].append(first_element)

# show the number of entries in the binary patterns file vs how many unique 
# fingerprints there are 
print "Number of entries:", len(lines - 1)
print "Number of unique fingerprints:", len(unique_fings)

# you could get the frequency of each of the fingerprints
# dump your fingerprints into a list of tuples 
fing_counts = [(fing, len(unique_fings[fing])) for fing in unique_fings]
# each tuple will contain the fingerprint and the count
# however, this list isn't sorted so to sort by descending order you can do the
# following:
fing_counts.sort(key=lambda tup: tup[1], reverse=True)
# This will sort the list by 2nd element in each tuple, the frequency count

# print how frequent each fingerprint is in the complete list of binary patterns
for fing, count in fing_counts:
	print "Fingerprint", fing, "appears", count, "times"