def wangzhangplot(fname):
seqlen = getseqlen(fname)
print(seqlen)
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
#x = Node(c)
seqarr.append(Node(c))
c = f.read(1)
f.close()
start_time = datetime.datetime.now()
'''print("non-A: ")
wangzhangcalc(seqarr, seqlen, 'A')
print("non-G: ")
wangzhangcalc(seqarr, seqlen, 'G')
print("non-C: ")
wangzhangcalc(seqarr, seqlen, 'C')'''
result = sqrt(
pow(wangzhangcalc(seqarr, seqlen, 'A'), 2) +
pow(wangzhangcalc(seqarr, seqlen, 'G'), 2) +
pow(wangzhangcalc(seqarr, seqlen, 'C'), 2))
print("mod: " + str(result))
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
def randic2dplot(fname):
window_list = list()
seqlen = getseqlen(fname)
print(seqlen, end=",")
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c!="":
c = c.upper()
if(c in baseset):
seqarr.append(Node(c))
c = f.read(1)
f.close()
for i in range (0, seqlen - 36, 3):
for j in range (i, i+36):
seqarr[j].reset()
# print("slide_count: " + str(i))
sliding_window(i, 36, seqarr, window_list)
# for window in window_list:
# print(str(window), end=",")
print(','.join(map(str, window_list)))
def taonanwangplot(fname):
'''f = open(fname, 'r')
seqlen=0
c = f.read(1)
while c!="":
if(c.upper() in baseset):
#print("string is: ", c)
seqlen+=1
c = f.read(1)
#print(seqlen)
f.close()'''
seqlen = getseqlen(fname)
start_time = datetime.datetime.now()
print("wsrocalc: ", wsrocalc(fname, seqlen))
print("mkrocalc: ", mkrocalc(fname, seqlen))
print("ryrocalc: ", ryrocalc(fname, seqlen))
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
def lifeizhaoyuplot(fname):
seqlen = getseqlen(fname)
#print(seqlen)
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
#x = Node(c)
seqarr.append(Node(c))
c = f.read(1)
f.close()
start_time = datetime.datetime.now()
point = basepos.O
count = 0
while (count < seqlen - 1):
word = seqarr[count].base + seqarr[count + 1].base
#print(count)
#print(word)
#(doubledict[word])
#point = point.unit_couplet_value(doubledict[word])
#point = doubledict[word].unit_couplet_value(point)
point = doubledict[word].add(point)
#print(point)
count += 1
#print(count)
print(point)
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
def nandyplot(fname):
seqlen = getseqlen(fname)
print(seqlen)
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
#x = Node(c)
seqarr.append(Node(c))
c = f.read(1)
f.close()
curr_x = 0
curr_y = 0
#curr_z = 0
count_a = 0
count_c = 0
count_g = 0
count_t = 0
sum_x = 0
sum_y = 0
total_count = 0
start_time = datetime.datetime.now()
edmat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
curr_base = seqarr[i].base
if (curr_base == 'G'):
total_count += 1
count_g += 1
curr_x = curr_x - 1
sum_x = sum_x + curr_x
sum_y = sum_y + curr_y
#curr_y = curr_y + 1
#curr_z = curr_z - 1
elif (curr_base == 'C'):
total_count += 1
count_c += 1
curr_y = curr_y + 1
sum_x = sum_x + curr_x
sum_y = sum_y + curr_y
#curr_z = curr_z + 1
elif (curr_base == 'A'):
total_count += 1
count_a += 1
curr_x = curr_x + 1
sum_x = sum_x + curr_x
sum_y = sum_y + curr_y
#curr_z = curr_z - 1
elif (curr_base == 'T'):
total_count += 1
count_t += 1
curr_y = curr_y - 1
sum_x = sum_x + curr_x
sum_y = sum_y + curr_y
#curr_z = curr_z + 1
seqarr[i].x = curr_x
seqarr[i].y = curr_y
mew_x = sum_x / total_count
mew_y = sum_y / total_count
gr_value = pow((pow(mew_x, 2.0) + pow(mew_y, 2.0)), 0.5)
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('gr_value = ' + str(gr_value))
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
def songtangplot(fname):
seqlen = getseqlen(fname)
print(seqlen)
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
#x = Node(c)
seqarr.append(Node(c))
c = f.read(1)
f.close()
curr_x = 0
curr_y = 0
start_time = datetime.datetime.now()
edmat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
'''if(seqarr[i].visited == False):
seqarr[i].x = i+1'''
for j in range(i, seqlen):
if (seqarr[j].visited == False):
curr_x += 1
seqarr[j].x = curr_x
seqarr[j].y = base_to_y_val_map[seqarr[j].base]
seqarr[j].visited = True
edmat[i][j] = sqrt(
pow((seqarr[i].x - seqarr[j].x), 2) +
pow((seqarr[i].y - seqarr[j].y), 2))
edmat[j][i] = edmat[i][j]
'''Computing the path-distance matrix'''
pdmat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (j == i):
pdmat[i][j] = 0
elif (i < j):
pdmat[i][j] = pdmat[i][j - 1] + edmat[j - 1][j]
pdmat[j][i] = pdmat[i][j]
'''Computing the M/M matrix'''
mbym_mat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (i == j):
mbym_mat[i][j] = 0
else:
mbym_mat[i][j] = edmat[i][j] / abs(i - j)
mbym_mat[j][i] = mbym_mat[i][j]
np.savetxt("songtang-mm.csv", mbym_mat, delimiter=",")
'''Computing the L/L matrix'''
lbyl_mat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (i == j):
lbyl_mat[i][j] = 0
else:
lbyl_mat[i][j] = edmat[i][j] / pdmat[i][j]
lbyl_mat[j][i] = lbyl_mat[i][j]
np.savetxt("songtang-ll.csv", lbyl_mat, delimiter=",")
print(np.linalg.eigvals(mbym_mat).max())
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
def randic3dplot(fname):
seqlen = getseqlen(fname)
print(seqlen)
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
#x = Node(c)
seqarr.append(Node(c))
c = f.read(1)
f.close()
curr_x = 0
curr_y = 0
curr_z = 0
start_time = datetime.datetime.now()
edmat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
if (seqarr[i].visited == False):
curr_base = seqarr[i].base
if (curr_base == 'G'):
curr_x = curr_x - 1
curr_y = curr_y + 1
curr_z = curr_z - 1
elif (curr_base == 'C'):
curr_x = curr_x - 1
curr_y = curr_y - 1
curr_z = curr_z + 1
elif (curr_base == 'A'):
curr_x = curr_x + 1
curr_y = curr_y - 1
curr_z = curr_z - 1
elif (curr_base == 'T'):
curr_x = curr_x + 1
curr_y = curr_y + 1
curr_z = curr_z + 1
seqarr[i].x = curr_x
seqarr[i].y = curr_y
seqarr[i].z = curr_z
seqarr[i].visited = True
for j in range(i, seqlen):
if (seqarr[j].visited == False):
curr_base = seqarr[j].base
if (curr_base == 'G'):
curr_x = curr_x - 1
curr_y = curr_y + 1
curr_z = curr_z - 1
elif (curr_base == 'C'):
curr_x = curr_x - 1
curr_y = curr_y - 1
curr_z = curr_z + 1
elif (curr_base == 'A'):
curr_x = curr_x + 1
curr_y = curr_y - 1
curr_z = curr_z - 1
elif (curr_base == 'T'):
curr_x = curr_x + 1
curr_y = curr_y + 1
curr_z = curr_z + 1
seqarr[j].x = curr_x
seqarr[j].y = curr_y
seqarr[j].z = curr_z
seqarr[j].visited = True
edmat[i][j] = sqrt(
pow((seqarr[i].x - seqarr[j].x), 2) +
pow((seqarr[i].y - seqarr[j].y), 2) +
pow((seqarr[i].z - seqarr[j].z), 2)) / pow(3, 0.5)
edmat[j][i] = edmat[i][j]
'''Computing the path-distance matrix'''
pdmat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (j == i):
pdmat[i][j] = 0
elif (i < j):
pdmat[i][j] = pdmat[i][j - 1] + edmat[j - 1][j]
pdmat[j][i] = pdmat[i][j]
'''Computing the M/M matrix'''
mbym_mat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (i == j):
mbym_mat[i][j] = 0
else:
mbym_mat[i][j] = edmat[i][j] / abs(i - j)
mbym_mat[j][i] = mbym_mat[i][j]
'''Computing the L/L matrix'''
lbyl_mat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (i == j):
lbyl_mat[i][j] = 0
else:
lbyl_mat[i][j] = edmat[i][j] / pdmat[i][j]
lbyl_mat[j][i] = lbyl_mat[i][j]
print(np.linalg.eigvals(mbym_mat).max())
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
def yauplot(fname):
seqlen = getseqlen(fname)
print(seqlen)
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
seqarr.append(Node(c))
c = f.read(1)
f.close()
curr_x = 0.0
curr_y = 0.0
count_a = 0
count_c = 0
count_g = 0
count_t = 0
sum_x = 0.0
sum_y = 0.0
total_count = 0
start_time = datetime.datetime.now()
sqrt_three = pow(3, 0.5)
half = 1.0 / 2.0
for i in range(0, seqlen):
curr_base = seqarr[i].base
if (curr_base == 'G'):
total_count += 1
count_g += 1
curr_x = float(curr_x + (sqrt_three * half))
sum_x = float(sum_x + curr_x)
curr_y = float(curr_y - half)
sum_y = float(sum_y + curr_y)
elif (curr_base == 'C'):
total_count += 1
count_c += 1
curr_x = float(curr_x + (sqrt_three * half))
sum_x = float(sum_x + curr_x)
curr_y = float(curr_y + half)
sum_y = float(sum_y + curr_y)
elif (curr_base == 'A'):
total_count += 1
count_a += 1
curr_x = float(curr_x + half)
sum_x = float(sum_x + curr_x)
curr_y = float(curr_y - (sqrt_three * half))
sum_y = float(sum_y + curr_y)
elif (curr_base == 'T'):
total_count += 1
count_t += 1
curr_x = float(curr_x + half)
sum_x = float(sum_x + curr_x)
curr_y = float(curr_y + (sqrt_three * half))
sum_y = float(sum_y + curr_y)
seqarr[i].x = curr_x
seqarr[i].y = curr_y
mew_x = float(sum_x / total_count)
mew_y = float(sum_y / total_count)
sqr_gr_value = pow(mew_x, 2.0) + pow(mew_y, 2.0)
gr_value = pow(sqr_gr_value, 0.5)
# gr_value = pow((pow(mew_x, 2.0) + pow(mew_y, 2.0)), 0.5)
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('mew_x = ' + str(mew_x))
print('mew_y = ' + str(mew_y))
print('sqr_gr_value = ' + str(sqr_gr_value))
print('gr_value = ' + str(gr_value))
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
def randic3dplot(fname):
seqlen = getseqlen(fname)
print("Current thread: " + str(threading.get_ident()) +
"; sequence length: " + str(seqlen))
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
#x = Node(c)
seqarr.append(Node(c))
c = f.read(1)
f.close()
start_time = datetime.datetime.now()
# edmat = np.zeros((seqlen, seqlen))
edmat = np.zeros((seqlen, seqlen), dtype=float)
# edmat = pd.DataFrame(np.zeros((seqlen, seqlen)))
for i in range(0, seqlen):
visit_node(seqarr[i])
for j in range(i, seqlen):
visit_node(seqarr[j])
edmat[i][j] = sqrt(
pow((seqarr[i].x - seqarr[j].x), 2) +
pow((seqarr[i].y - seqarr[j].y), 2) +
pow((seqarr[i].z - seqarr[j].z), 2)) / pow(3, 0.5)
edmat[j][i] = edmat[i][j]
# pool = Pool(processes=2)
# for i in range(0, seqlen):
# for j in range(i, seqlen):
# #pool.map(memoize_ed(seqarr, edmat, i, j))
# pool.map(memoize_ed, seqarr)
# '''Computing the path-distance matrix'''
# pdmat = np.zeros((seqlen, seqlen))
# for i in range(0, seqlen):
# for j in range(i, seqlen):
# if(j==i):
# pdmat[i][j] = 0
# elif(i<j):
# pdmat[i][j] = pdmat[i][j-1] + edmat[j-1][j]
# pdmat[j][i]= pdmat[i][j]
# '''Computing the M/M matrix'''
# mbym_mat = np.zeros((seqlen, seqlen))
# print(current.name, current._identity + " Size of edmat:" + str(edmat.nbytes))
# print("Current thread: " + str(threading.current_thread().getName) + "; Size of edmat(MB):" + str(edmat.nbytes/(1024*1024)))
# print("Current thread: " + str(threading.Thread.getName) + "; Size of edmat(MB):" + str(edmat.nbytes/(1024*1024)))
print("Current thread: " + str(threading.get_ident()) +
"; Size of edmat(MB):" + str(edmat.nbytes / (1024 * 1024)))
# mbym_mat = pd.DataFrame(np.zeros((seqlen, seqlen)))
# mbym_mat = np.zeros((seqlen, seqlen))
mbym_mat = np.zeros((seqlen, seqlen), dtype=float)
for i in range(0, seqlen):
for j in range(i, seqlen):
if (i == j):
mbym_mat[i][j] = 0
else:
mbym_mat[i][j] = edmat[i][j] / abs(i - j)
mbym_mat[j][i] = mbym_mat[i][j]
# print(current.name, current._identity + " Size of mbym_mat:" + str(mbym_mat.nbytes))
print("Current thread: " + str(threading.get_ident()) +
"; Size of mbym_mat(MB):" + str(mbym_mat.nbytes / (1024 * 1024)))
# '''Computing the L/L matrix'''
# lbyl_mat = np.zeros((seqlen, seqlen))
# for i in range(0, seqlen):
# for j in range(i, seqlen):
# if(i==j):
# lbyl_mat[i][j] = 0;
# else:
# lbyl_mat[i][j] = edmat[i][j] / pdmat[i][j]
# lbyl_mat[j][i] = lbyl_mat[i][j]
# print(current.name, current._identity + " Max eigvalue: " + np.linalg.eigvals(mbym_mat).max())
# print("Max eigvalue: " + np.linalg.eigvals(mbym_mat).max())
before_eigvals = datetime.datetime.now()
# print("Max eigvalue(eig): " + str(np.linalg.eigvals(mbym_mat).max()) + ": time_taken: " + str(datetime.datetime.now() - before_eigvals))
before_eigvalsh = datetime.datetime.now()
print("Current thread: " + str(threading.get_ident()) +
"; Max eigvalue(eigh): " + str(np.linalg.eigvalsh(mbym_mat).max()) +
": time_taken: " + str(datetime.datetime.now() - before_eigvalsh))
end_time = datetime.datetime.now()
time_taken = end_time - start_time
# print('time taken = ' + str(time_taken.microseconds))
# print("Current thread: " + str(threading.get_ident()) + "; time taken = " + str(time_taken.seconds))
print("Current thread: " + str(threading.get_ident()) + "; time taken = " +
str(time_taken))
gc.collect()
def jiliplot(fname):
seqlen = getseqlen(fname)
print(seqlen)
seqarr = []
f = open(fname, 'r')
c = f.read(1)
while c != "":
c = c.upper()
if (c in baseset):
#x = Node(c)
seqarr.append(Node(c))
c = f.read(1)
f.close()
#print(seqarr)
#f = open(fname, 'r')
#c = f.read(1).upper()
rcount = 0
ycount = 0
start_time = datetime.datetime.now()
'''Computing the Euclidean-distance-distance matrix'''
edmat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
#seqarr[i].x = 0; seqarr[i].y = 0;
if (seqarr[i].base in baseset):
if (seqarr[i].visited == False):
if (seqarr[i].base in rset):
seqarr[i].x = 1
rcount += 1
seqarr[i].y = rcount
else:
seqarr[i].x = 0
ycount += 1
seqarr[i].y = ycount
seqarr[i].visited = True
for j in range(i, seqlen):
#seqarr[j].x = 0; seqarr[j].y = 0;
if (j != i):
if (seqarr[j].base in baseset):
if (seqarr[j].visited == False):
if (seqarr[j].base in rset):
seqarr[j].x = 1
rcount += 1
seqarr[j].y = rcount
else:
seqarr[j].x = 0
ycount += 1
seqarr[j].y = ycount
seqarr[j].visited = True
else:
seqarr[j].x = seqarr[i].x
seqarr[j].y = seqarr[i].y
edmat[i][j] = sqrt(
pow((seqarr[i].x - seqarr[j].x), 2) +
pow((seqarr[i].y - seqarr[j].y), 2))
edmat[j][i] = edmat[i][j]
'''Computing the path-distance matrix'''
pdmat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (j == i):
pdmat[i][j] = 0
elif (i < j):
pdmat[i][j] = pdmat[i][j - 1] + edmat[j - 1][j]
pdmat[j][i] = pdmat[i][j]
'''Computing the M/M matrix'''
mbym_mat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (i == j):
mbym_mat[i][j] = 0
else:
mbym_mat[i][j] = edmat[i][j] / abs(i - j)
mbym_mat[j][i] = mbym_mat[i][j]
'''Computing the L/L matrix'''
lbyl_mat = np.zeros((seqlen, seqlen))
for i in range(0, seqlen):
for j in range(i, seqlen):
if (i == j):
lbyl_mat[i][j] = 0
else:
lbyl_mat[i][j] = edmat[i][j] / pdmat[i][j]
lbyl_mat[j][i] = lbyl_mat[i][j]
#print(pdmat)
#print('time taken = ' + str(time_taken.microseconds))
#print('time taken = ' + str(time_taken.seconds))
'''print(np.linalg.eigvals(edmat).max())
print(np.linalg.eigvals(pdmat).max())
print(np.linalg.eigvals(mbym_mat).max())
print(np.linalg.eigvals(lbyl_mat).max())'''
''' open('jili_ed.csv', 'w', newline='') as csvfile:
writer = csv.writer(csvfile, delimiter=',',
quotechar='|', quoting=csv.QUOTE_MINIMAL)
writer.writerows(edmat)
with open('jili_mm.csv', 'w', newline='') as csvfile:
writer = csv.writer(csvfile, delimiter=',',
quotechar='|', quoting=csv.QUOTE_MINIMAL)
writer.writerows(mbym_mat)
with open('jili_ll.csv', 'w', newline='') as csvfile:
writer = csv.writer(csvfile, delimiter=',',
quotechar='|', quoting=csv.QUOTE_MINIMAL)
writer.writerows(lbyl_mat)
with open('jili_pd.csv', 'w', newline='') as csvfile:
writer = csv.writer(csvfile, delimiter=',',
quotechar='|', quoting=csv.QUOTE_MINIMAL)
writer.writerows(pdmat)'''
'''print('**************************************************')
print(edmat)
print('**************************************************')
print('**************************************************')
print(mbym_mat)
print('**************************************************')
print('**************************************************')
print(lbyl_mat)
print('**************************************************')'''
print(np.linalg.eigvals(lbyl_mat).max())
end_time = datetime.datetime.now()
time_taken = end_time - start_time
print('time taken = ' + str(time_taken.microseconds))
print('time taken = ' + str(time_taken.seconds))
