import random
import math

#gcplus model
gcplus = [[]]
gcplus[0] = [0.18, 0.274, 0.426, 0.12]
gcplus.append([0.172, 0.368, 0.274, 0.188])
gcplus.append([0.161, 0.339, 0.375, 0.125])
gcplus.append([0.079, 0.355, 0.384, 0.182])
# a different model
gcminus = [[]]
gcminus[0] = [0.1, 0.1, 0.1, 0.7]
gcminus.append([0.1, 0.1, 0.7, 0.1])
gcminus.append([0.1, 0.7, 0.1, 0.1])
gcminus.append([0.7, 0.1, 0.1, 0.1])

# map symbols to their columns
dictionary = {'a':0, 'c':1, 'g':2, 't':3}

#generate a string of length L using a model
def generate(L):
    current = 'a' # current state
    s = '' # generated string
    for i in range(0,L):
        summ = 0
        r = random.random()  # by default 0-1
        for c in range(0,4):  # for each column in the model
            summ = summ + gcminus[dictionary[current]][c] # sum the probabilities
            if(r < summ):  # if the random value is less then the sum use that column
                if(c == 0):  # based on the column pick the state/symbol
                    current = 'a'
                if(c == 1):
                    current = 'c'
                if(c == 2):
                    current = 'g'
                if(c == 3):
                    current = 't'
                print(current, end="")
                break

# calculate the log-odds ratio of two models for a given string      
def compare(x):
    s = 0
    for i in range(1,len(x)):
        aplus = gcplus[dictionary[x[i-1]]][dictionary[x[i]]]
        aminus = gcminus[dictionary[x[i-1]]][dictionary[x[i]]]
        s = s + math.log(aplus/aminus,2)
    print(s)