clumpFinder.py

# print(loadFile('./Downloads/Vibrio_cholerae.txt', 'CTTGATCAT'))
  
"""
FindClumps(Text, k, L, t)
    Patterns ← an array of strings of length 0
    n ← |Text|
    for every integer i between 0 and n − L
        Window ← Text(i, L)
        freqMap ← FrequencyTable(Window, k)
        for every key s in freqMap
            if freqMap[s] ≥ t
                append s to Patterns
    remove duplicates from Patterns
    return Patterns
}
"""
def getFreqMap(text, k):
  frequencyMap = {}
  textLength = len(text)
  for i in range(0, (textLength-k)+1):
    pattern = text[i:(i+k)]
    if pattern in frequencyMap:
      frequencyMap[pattern] = frequencyMap[pattern] + 1
    else:
      frequencyMap[pattern] = 1
  return frequencyMap

def findClumps(genome, k, windowLength, leastFrequency):
  patterns = []
  genomeLength = len(genome)
  print(genomeLength)
  for i in range(0, (genomeLength-windowLength)+1):
    window = genome[i: i + windowLength]
    freqMap = getFreqMap(window, k)
   
    for key in freqMap:
      if freqMap[key] >= leastFrequency:
        patterns.append(key)
  return 0 if len(patterns) < 1 else list(set(patterns))

k = 5
windowLength = 50
leastFrequency = 4

def loadFile(path, pattern):
  file = open(path, 'r')
  genome = "".join(file.readlines())
  return findClumps(genome, k, windowLength, leastFrequency)

print(findClumps(genome, k, windowLength, leastFrequency))