def get_RSA_frequencies(natural_proteins, lower_RSA_boundary, upper_RSA_boundary):
natural_distribution = af.get_AA_distribution(natural_proteins)
natural_RSA = af.get_RSA_Values(natural_proteins)
natural_RSA_array = af.make_array(natural_RSA)
seq_length = len(natural_RSA)
bin_1 = []
bin_2 = []
bin_3 = []
bin_4 = []
bin_5 = []
i = 0
count = 0
for site in natural_distribution:
if (lower_RSA_boundary<=natural_RSA_array[i] and natural_RSA_array[i]<= upper_RSA_boundary):
#print natural_RSA_array[i]
#print site[0:4]
bin_1.append(site[0])
bin_2.append(site[1])
bin_3.append(site[2])
bin_4.append(site[3])
bin_5.append(site[4])
i = i + 1
count = count + 1
else:
i = i + 1
if count == 0:
frequency_data = [0.0, 0.0, 0.0, 0.0, 0.0]
else:
frequency_data = [mean(bin_1)/mean(bin_1), mean(bin_2)/mean(bin_1), mean(bin_3)/mean(bin_1), mean(bin_4)/mean(bin_1), mean(bin_5)/mean(bin_1)]
if (mean(bin_1)) == 0.0:
print "MEAN OF BIN 1 is ZERO!!!!"
print frequency_data
#frequency_data = [mean(bin_1), mean(bin_2), mean(bin_3), mean(bin_4), mean(bin_5)]
print "Number of residues in bin: " + str(count)
return frequency_data
def get_position_count_data(file):
buried_counts = []
surface_counts = []
buried_sites = []
surface_sites= []
input = open(file, "r")
protein_data = input.readlines()
input.close()
RSA = af.get_RSA_Values(file)
alignment_length = len(RSA)
site_data = af.get_transformed_data(protein_data)
#print RSA
#print site_data
i = 0
for site in site_data:
if(float(RSA[i]) < 0.05):
buried_sites.append(site)
else:
surface_sites.append(site)
i = i + 1
buried_array = np.array(buried_sites)
surface_array = np.array(surface_sites)
#print buried_array
#print buried_array
buried_m, buried_n = buried_array.shape
buried_total_sum = sum(sum(buried_array))
#print buried_m
#print buried_n
surface_m, surface_n = surface_array.shape
surface_total_sum = sum(sum(surface_array))
#print surface_total_sum
j = 0
while (j < buried_n):
buried_site_sum = sum(buried_array[:, j])
buried_counts.append(float(buried_site_sum)/float(buried_total_sum))
j = j + 1
j = 0
while (j < surface_n):
surface_site_sum = sum(surface_array[:, j])
surface_counts.append(float(surface_site_sum)/float(surface_total_sum))
j = j + 1
return buried_counts, surface_counts
def get_mixed_entropy_values(PDB, buried_temp, surface_temp):
new_entropies = []
#Make the files
buried_file = "align_data_array_" + PDB + "_" + str(buried_temp) + ".dat"
surface_file = "align_data_array_" + PDB + "_" + str(surface_temp) + ".dat"
#Get the RSA Values
RSA = af.make_array(af.get_RSA_Values(buried_file))
buried_entropies = af.get_native_entropy(buried_file)
surface_entropies = af.get_native_entropy(surface_file)
#Get the entropy values
for i in xrange(len(RSA)):
if (float(RSA[i]) <=0.25):
new_entropies.append(buried_entropies[i])
else:
new_entropies.append(surface_entropies[i])
return RSA, new_entropies
def get_position_count_data(file):
buried_counts = []
surface_counts = []
buried_sites = []
surface_sites = []
input = open(file, "r")
protein_data = input.readlines()
input.close()
RSA = af.get_RSA_Values(file)
alignment_length = len(RSA)
site_data = af.get_transformed_data(protein_data)
i = 0
for site in site_data:
if (float(RSA[i]) <= 0.05):
buried_sites.append(site)
else:
surface_sites.append(site)
i = i + 1
buried_array = np.array(buried_sites)
surface_array = np.array(surface_sites)
buried_m, buried_n = buried_array.shape
buried_total_sum = sum(sum(buried_array))
surface_m, surface_n = surface_array.shape
surface_total_sum = sum(sum(surface_array))
#print surface_total_sum
j = 0
while (j < buried_n):
buried_site_sum = sum(buried_array[:, j])
buried_counts.append(float(buried_site_sum) / float(buried_total_sum))
j = j + 1
j = 0
while (j < surface_n):
surface_site_sum = sum(surface_array[:, j])
surface_counts.append(
float(surface_site_sum) / float(surface_total_sum))
j = j + 1
return buried_counts, surface_counts
def get_RSA_frequencies(natural_proteins, lower_RSA_boundary, upper_RSA_boundary):
#natural_distribution = af.get_AA_distribution(natural_proteins)
natural_distribution = get_AA_distribution_mod(natural_proteins)
#natural_dis_array = array(natural_distribution)
#m,n = natural_dis_array.shape
#print "num_residues, length of alignment: " + str(n),m
#print natural_distribution
natural_RSA = af.get_RSA_Values(natural_proteins)
natural_RSA_array = af.make_array(natural_RSA)
seq_length = len(natural_RSA)
frequency_data = []
all_k_values = []
k_values = []
k_lists = []
bin_1 = []
bin_2 = []
bin_3 = []
bin_4 = []
bin_5 = []
bin_6 = []
bin_7 = []
bin_8 = []
bin_9 = []
bin_10 = []
bin_11 = []
bin_12 = []
bin_13 = []
bin_14 = []
bin_15 = []
bin_16 = []
bin_17 = []
bin_18 = []
bin_19 = []
bin_20 = []
i = 0
count = 0
for site in natural_distribution:
if (lower_RSA_boundary<=natural_RSA_array[i] and natural_RSA_array[i]<= upper_RSA_boundary):
#print natural_RSA_array[i]
#print site[0:4]
bin_1.append(site[0])
bin_2.append(site[1])
bin_3.append(site[2])
bin_4.append(site[3])
bin_5.append(site[4])
bin_6.append(site[5])
bin_7.append(site[6])
bin_8.append(site[7])
bin_9.append(site[8])
bin_10.append(site[9])
bin_11.append(site[10])
bin_12.append(site[11])
bin_13.append(site[12])
bin_14.append(site[13])
bin_15.append(site[14])
bin_16.append(site[15])
bin_17.append(site[16])
bin_18.append(site[17])
bin_19.append(site[18])
bin_20.append(site[19])
i = i + 1
count = count + 1
else:
i = i + 1
if count != 0: #Need to find a way to exclude the point
frequencies = [np.mean(bin_1)/np.mean(bin_1), np.mean(bin_2)/np.mean(bin_1), np.mean(bin_3)/np.mean(bin_1), np.mean(bin_4)/np.mean(bin_1), np.mean(bin_5)/np.mean(bin_1), np.mean(bin_6)/np.mean(bin_1), np.mean(bin_7)/np.mean(bin_1), np.mean(bin_8)/np.mean(bin_1), np.mean(bin_9)/np.mean(bin_1), np.mean(bin_10)/np.mean(bin_1), np.mean(bin_11)/np.mean(bin_1), np.mean(bin_12)/np.mean(bin_1), np.mean(bin_13)/np.mean(bin_1), np.mean(bin_14)/np.mean(bin_1), np.mean(bin_15)/np.mean(bin_1), np.mean(bin_16)/np.mean(bin_1), np.mean(bin_17)/np.mean(bin_1), np.mean(bin_18)/np.mean(bin_1), np.mean(bin_19)/np.mean(bin_1), np.mean(bin_20)/np.mean(bin_1)]
all_k_values = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19]
k_count = 0
for element in frequencies:
if element == 0.0:
continue
else:
frequency_data.append(np.log(element))
k_values.append(all_k_values[k_count])
k_count = k_count + 1
set_k_r(k_values, lower_RSA_boundary, upper_RSA_boundary)
return frequency_data
count = 0
for temp in temps:
buried_percentage_list = []
for pdb_id in PDBS:
file = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(temp) + ".dat"
#fileparts = re.split("_",file)
#pdb_id = fileparts[3].upper() #Gets the PDB Name and the chain_id
#chain_id = fileparts[6]
#chain_id = chain_id[0]
#print "Processing file: " + file
#pdb_names.append(pdb_id)
#chain_names.append(chain_id)
proteins = file
RSA = af.get_RSA_Values(proteins)
#natural_RSA_array = af.make_array(natural_RSA)
buried_percentage = get_percent_buried(RSA)
buried_percentage_list.append(buried_percentage)
#print temp, RSA
#print str(temp), buried_percentage_list, "/n"
N = 40 #The number of bars in each group - should be 20 for the 20 amino acids
index = np.arange(N) # The x locations for the groups
fig = plt.figure(count)
ax = plt.axes([0.07, 0.14, 0.90, 0.83])
width = 0.40 #The width of the bars
b1 = plt.bar(index, buried_percentage_list, width, color = "blue")
temp_string = "Temp = " + str(temp)
ax.text(37.5, 0.325, temp_string, fontweight = 'bold', ha = 'center', fontsize = 14)#, va = 'center', fontsize = 16)
#ax.text(0.6, 0.265,"Temp = " + str(temp), fontweight = 'bold', fontsize = 18)#, ha = 'center', va = 'center', fontsize = 16)
for PDB in PDBS:
RSA1, entropy_mix1 = get_mixed_entropy_values(PDB, 0.0, 0.1)
RSA2, entropy_mix2 = get_mixed_entropy_values(PDB, 0.03, 0.1)
[cor_entropy_RSA_mix1, pvalue1] = pearsonr(RSA1, entropy_mix1)
cor_entropy_RSA_mix1 = float(cor_entropy_RSA_mix1)
cor_values1.append(cor_entropy_RSA_mix1)
[cor_entropy_RSA_mix2, pvalue2] = pearsonr(RSA2, entropy_mix2)
cor_entropy_RSA_mix2 = float(cor_entropy_RSA_mix2)
cor_values2.append(cor_entropy_RSA_mix2)
natural_file = "align_natural_data_array_" + PDB + ".dat"
natural_RSA = af.make_array(af.get_RSA_Values(natural_file))
natural_entropy = af.get_native_entropy(natural_file)
[natural_cor_entropy_RSA, pvalue3] = pearsonr(natural_RSA, natural_entropy)
natural_cor_entropy_RSA = float(natural_cor_entropy_RSA)
natural_cor_values.append(natural_cor_entropy_RSA)
fig = plt.figure(1, dpi = 400, figsize = (16,6))
correlation_values = [cor_values1, cor_values2, natural_cor_values]
correlation_values_transpose = transpose(correlation_values)
(m,n) = correlation_values_transpose.shape
#rcParams['lines.linewidth'] = 2
ax = axes([0.066, 0.115, 0.43, 0.85])
#text(-0.37, 0.6, "A", fontweight = 'bold', ha = 'center', va = 'center', fontsize = 20)
'''
count = 0
for temp in temps:
buried_percentage_list = []
for pdb_id in PDBS:
file = "align_data_array_" + pdb_id + "_" + chain_id + "_" + "soft.dat"
#fileparts = re.split("_",file)
#pdb_id = fileparts[3].upper() #Gets the PDB Name and the chain_id
#chain_id = fileparts[6]
#chain_id = chain_id[0]
print "Processing file: " + file
#pdb_names.append(pdb_id)
#chain_names.append(chain_id)
natural_proteins = file
natural_RSA = af.get_RSA_Values(natural_proteins)
#natural_RSA_array = af.make_array(natural_RSA)
buried_percentage = get_percent_buried(natural_RSA)
buried_percentage_list.append(buried_percentage)
print buried_percentage_list
N = 40 #The number of bars in each group - should be 20 for the 20 amino acids
index = np.arange(N) # The x locations for the groups
fig = plt.figure(count)
ax = plt.axes([0.07, 0.14, 0.90, 0.83])
width = 0.40 #The width of the bars
b1 = plt.bar(index, buried_percentage_list, width, color = "blue")
temp_string = "Temp = " + str(temp)
ax.text(37.5, 0.325, temp_string, fontweight = 'bold', ha = 'center', fontsize = 14)#, va = 'center', fontsize = 16)
#ax.text(0.6, 0.265,"Temp = " + str(temp), fontweight = 'bold', fontsize = 18)#, ha = 'center', va = 'center', fontsize = 16)
ax.set_xticklabels(PDBS, rotation = 'vertical')
designed_proteins_rosetta = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(
"rosetta") + ".dat"
designed_proteins_evolved = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(
"evolved") + ".dat"
split_natural_1 = "align_natural_sample1_data_array_" + pdb_id + "_" + chain_id + ".dat"
split_natural_2 = "align_natural_sample2_data_array_" + pdb_id + "_" + chain_id + ".dat"
#Calculates all of the data for comparison (ex. entropy)
natural_distribution = analysis_functions.get_AA_distribution_KL(
natural_proteins)
natural_entropy = analysis_functions.get_native_entropy(
natural_proteins)
natural_entropy_array = analysis_functions.make_array(
natural_entropy)
natural_RSA = analysis_functions.get_RSA_Values(natural_proteins)
natural_RSA_array = analysis_functions.make_array(natural_RSA)
natural_mean_RSA_values.append(mean(natural_RSA_array))
natural_mean_entropy_values.append(mean(natural_entropy_array))
#Calculates cn & wcn
# cn13_data = analysis_functions.get_cn13_values(pdb_id, chain_id)
# iCN13 = cn13_data[0]
# iCN13_array = analysis_functions.make_array(cn13_data)
# mean_iCN13_values.append(mean(iCN13_array))
iwcn_data = calc_wcn.get_iwcn_values(pdb_id, chain_id)
iWCN_array = analysis_functions.make_array(iwcn_data)
mean_iWCN_values.append(mean(iWCN_array))
designed_distribution_rosetta = analysis_functions.get_AA_distribution_KL(
def get_RSA_frequencies(natural_proteins, lower_RSA_boundary, upper_RSA_boundary):
natural_distribution = af.get_AA_distribution(natural_proteins)
#natural_distribution = get_AA_distribution_mod(natural_proteins)
#natural_dis_array = array(natural_distribution)
#m,n = natural_dis_array.shape
#print "num_residues, length of alignment: " + str(n),m
#print natural_distribution
natural_RSA = af.get_RSA_Values(natural_proteins)
natural_RSA_array = af.make_array(natural_RSA)
seq_length = len(natural_RSA)
frequency_data = []
bin_1 = []
bin_2 = []
bin_3 = []
bin_4 = []
bin_5 = []
bin_6 = []
bin_7 = []
bin_8 = []
bin_9 = []
bin_10 = []
bin_11 = []
bin_12 = []
bin_13 = []
bin_14 = []
bin_15 = []
bin_16 = []
bin_17 = []
bin_18 = []
bin_19 = []
bin_20 = []
i = 0
count = 0
for site in natural_distribution:
if (lower_RSA_boundary<=natural_RSA_array[i] and natural_RSA_array[i]<= upper_RSA_boundary):
#print natural_RSA_array[i]
#print site[0:4]
bin_1.append(site[0])
bin_2.append(site[1])
bin_3.append(site[2])
bin_4.append(site[3])
bin_5.append(site[4])
bin_6.append(site[5])
bin_7.append(site[6])
bin_8.append(site[7])
bin_9.append(site[8])
bin_10.append(site[9])
bin_11.append(site[10])
bin_12.append(site[11])
bin_13.append(site[12])
bin_14.append(site[13])
bin_15.append(site[14])
bin_16.append(site[15])
bin_17.append(site[16])
bin_18.append(site[17])
bin_19.append(site[18])
bin_20.append(site[19])
i = i + 1
count = count + 1
else:
i = i + 1
if count == 0: #Need to find a way to exclude the point
frequency_data = [-1] # [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
#return frequency_data
else:
frequencies = [np.mean(bin_1)/np.mean(bin_1), np.mean(bin_2)/np.mean(bin_1), np.mean(bin_3)/np.mean(bin_1), np.mean(bin_4)/np.mean(bin_1), np.mean(bin_5)/np.mean(bin_1), np.mean(bin_6)/np.mean(bin_1), np.mean(bin_7)/np.mean(bin_1), np.mean(bin_8)/np.mean(bin_1), np.mean(bin_9)/np.mean(bin_1), np.mean(bin_10)/np.mean(bin_1), np.mean(bin_11)/np.mean(bin_1), np.mean(bin_12)/np.mean(bin_1), np.mean(bin_13)/np.mean(bin_1), np.mean(bin_14)/np.mean(bin_1), np.mean(bin_15)/np.mean(bin_1), np.mean(bin_16)/np.mean(bin_1), np.mean(bin_17)/np.mean(bin_1), np.mean(bin_18)/np.mean(bin_1), np.mean(bin_19)/np.mean(bin_1), np.mean(bin_20)/np.mean(bin_1)]
for element in frequencies:
if element == 0.0:
frequency_data.append(0.0)
else:
frequency_data.append(np.log(element))
#print np.log(element)
#if (mean(bin_1)) == 0.0:
# print "MEAN OF BIN 1 is ZERO!!!!"
#print frequency_data
#frequency_data = [mean(bin_1), mean(bin_2), mean(bin_3), mean(bin_4), mean(bin_5)]
#print "Number of residues in bin: " + str(count)
return frequency_data
designed_proteins_00 = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(0.0) + ".dat" designed_proteins_01 = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(0.1) + ".dat" designed_proteins_03 = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(0.3) + ".dat" designed_proteins_06 = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(0.6) + ".dat" designed_proteins_09 = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(0.9) + ".dat" designed_proteins_12 = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(1.2) + ".dat" designed_proteins_003 = "align_data_array_" + pdb_id + "_" + chain_id + "_" + str(0.03) + ".dat" split_natural_1 = "align_natural_sample1_data_array_" + pdb_id + "_" + chain_id + ".dat" split_natural_2 = "align_natural_sample2_data_array_" + pdb_id + "_" + chain_id + ".dat" #Calculates all of the data for comparison (ex. entropy) natural_distribution = analysis_functions.get_AA_distribution(natural_proteins) natural_entropy = analysis_functions.get_native_entropy(natural_proteins) natural_entropy_array = analysis_functions.make_array(natural_entropy) natural_RSA = analysis_functions.get_RSA_Values(natural_proteins) natural_RSA_array = analysis_functions.make_array(natural_RSA) natural_mean_RSA_values.append(mean(natural_RSA_array)) natural_mean_entropy_values.append(mean(natural_entropy_array)) designed_distribution_00 = analysis_functions.get_AA_distribution(designed_proteins_00) designed_entropy_00 = analysis_functions.get_native_entropy(designed_proteins_00) designed_entropy_array_00 = analysis_functions.make_array(designed_entropy_00) designed_RSA_00 = analysis_functions.get_RSA_Values(designed_proteins_00) designed_RSA_array_00 = analysis_functions.make_array(designed_RSA_00) designed_mean_RSA_values_00.append(mean(designed_RSA_array_00)) designed_mean_entropy_values_00.append(mean(designed_entropy_array_00)) designed_distribution_01 = analysis_functions.get_AA_distribution(designed_proteins_01)
