Local alignment of sequence

Aceasta postare este o continuare la Global alignment of DNA sequence, cu diferenta ca matricea de valori este initializata toata cu 0 (inclusiv marginile), iar valorile negative din matricea de costuri sunt rotunjite la 0.
Am considerat costurile:  match = 2, mismatch = -1, gap = -2

Prima matrice: costurile
A doua matrice: precedenta - 0 (diagonala), 1 (vine de sus), 2 (vine din stanga), -1 (neinitializat)

Sursa se afla aici.

Global alignment of DNA sequence (Python)

Programul gaseste o aliniere intre doua secvente {acgt} astfel incat castigul sa fie maxim.
Costuri considerate: match = 1 ; mismatch = -1 ; gap = -2

Exemplu de rulare:

Sursa se afla aici.

Amintiri din Poli

Acel moment cand popcorn intarzie 55 min la examen, vine total nepregatit si incearca fara succes sa scoata niste subiecte din cap, dureaza 10 min sa dicteze o intrebare grila + 4 raspunsuri dintre care te prinzi care e ala corect pt ca il zice instant si la alea gresite se gandeste cate 1 min gen : "ĂĂĂĂĂ, MMMMM" -- via Facebook

From DNA to Protein

Perceptron AND/OR gate

threshold = 0.5 # any value
learning_rate = 0.1 # any value
weights = [1, 1, 1] # x0 = 1 is constant, x1, x2 any value

# AND GATE
training_set = [((1, 0, 0), 0), ((1, 0, 1), 0), ((1, 1, 0), 0), ((1, 1, 1), 1)]

# OR GATE
#training_set = [((1, 0, 0), 0), ((1, 0, 1), 1), ((1, 1, 0), 1), ((1, 1, 1), 1)]

def dot_product(values, weights):
return sum(value * weight for value, weight in zip(values, weights))

while True:
    print('-' * 60)
    error_count = 0
    for input_vector, desired_output in training_set:
        result = dot_product(input_vector, weights) > threshold
        error = desired_output - result
        if error != 0:
            error_count += 1
            for index, value in enumerate(input_vector):
                weights[index] += learning_rate * error * value
    if error_count == 0:
        break

print(str(weights[1]) + ", " + str(weights[2])) # w1 & w2
print(threshold-weights[0]) # threshold