def map_ag_for_proposed_algo(antigen):
alleles = ag_to_allele_dict[antigen]
set_alleles = []
for allele in alleles:
new_allele = allele_truncate(allele)
set_alleles.append(new_allele)
alleles_mapped = list(set(set_alleles))
return alleles_mapped
def allele_list_ags(allele_list, pop):
ag_freq = 0.0
allele_antigen_freq = {}
for allele in allele_list:
allele = allele.split("+")[0]
allele = allele.rstrip("g p P G")
allele = hla.allele_truncate(allele)
if allele in population_allele_frequencies[pop]:
ag_freq = population_allele_frequencies[pop][allele]
else:
ag_freq = 0.0
ag = allele_to_ag_dict[allele][0]
bw46 = allele_to_ag_dict[allele][2]
allele_antigen = ag
if allele_antigen in allele_antigen_freq.keys():
allele_antigen_freq[allele_antigen] += float(ag_freq)
else:
allele_antigen_freq[allele_antigen] = float(ag_freq)
print (allele + " " + ag + " " + str(ag_freq))
TF = sum(allele_antigen_freq.values())
if TF == 0.0:
TF = 1
else:
TF = TF
for i,j in allele_antigen_freq.items():
ag_probs = j/TF
#ag_probs = round(ag_probs, 4)
allele_antigen_freq[i] = ag_probs
sorted_af = sorted(allele_antigen_freq.items(), key = operator.itemgetter(1), reverse = True)
top_ag = sorted_af[0][0]
top_af = sorted_af[0][1]
return (top_ag, bw46, sorted_af)
ua_ag = row_split[0]
ua_ag_eqs = row_split[1:]
ua_ag_eqs = list(filter(None, ua_ag_eqs))
UA_eq_dict[ua_ag] = ua_ag_eqs
#print(UA_eq_dict)
UNOS_conversion_table_filename = "conversion_table.csv"
UNOS_conversion_table_file = open(UNOS_conversion_table_filename, 'r')
for row in UNOS_conversion_table_file:
expression_character = ""
if row.startswith("Allele"):
continue
else:
allele = row.split(',')[0]
allele_4d = hla.allele_truncate(allele)
antigen = row.split(',')[1]
if antigen in ag_to_allele_dict.keys():
if allele_4d in ag_to_allele_dict[antigen]:
continue
else:
ag_to_allele_dict[antigen].append(allele_4d)
else:
ag_to_allele_dict[antigen] = [allele_4d]
#print(ag_to_allele_dict)
final_dict = {}
def genotype_ags(genotype_list, pop):
ag_freq_1 = 0.0
ag_freq_2 = 0.0
geno_antigen_freq = {}
for genotype in genotype_list:
allele_1 = genotype.split("+")[0]
allele_1 = allele_1.rstrip("g")
allele_1 = hla.allele_truncate(allele_1)
allele_2 = genotype.split("+")[1]
allele_2 = allele_2.rstrip("g")
allele_2 = hla.allele_truncate(allele_2)
ag_1 = allele_to_ag_dict[allele_1][0]
bw46_1 = allele_to_ag_dict[allele_1][2]
ag_2 = allele_to_ag_dict[allele_2][0]
bw46_2 = allele_to_ag_dict[allele_2][2]
if allele_1 in population_allele_frequencies[pop]:
ag_freq_1 = population_allele_frequencies[pop][allele_1]
if allele_2 in population_allele_frequencies[pop]:
ag_freq_2 = population_allele_frequencies[pop][allele_2]
gf = 0
if (ag_1 == ag_2):
gf = float(ag_freq_1) * float(ag_freq_2)
else:
gf = 2 * float(ag_freq_1) * float(ag_freq_2)
geno_antigen = ag_1 + "+" + ag_2
if geno_antigen in geno_antigen_freq.keys():
geno_antigen_freq[geno_antigen] += float(gf)
else:
geno_antigen_freq[geno_antigen] = float(gf)
TF = sum(geno_antigen_freq.values())
if TF == 0.0:
TF = 1
else:
TF = TF
for i, j in geno_antigen_freq.items():
ag_probs = j / TF
geno_antigen_freq[i] = ag_probs
print(geno_antigen_freq)
sorted_gf = sorted(geno_antigen_freq.items(),
key=operator.itemgetter(1),
reverse=True)
print(sorted_gf)
#if len(sorted_gf) == 1:
#ag_prob = 1
#else:
#antigen_list = sortef_gf[1::2]
top_ag_geno = sorted_gf[0][0]
top_gf = sorted_gf[0][1]
#print(top_ag_geno)
ag_1 = top_ag_geno.split("+")[0]
ag_2 = top_ag_geno.split("+")[1]
#print(ag_1)
#print(ag_2)
ag_list = ag_1 + "," + ag_2
bw46_list = bw46_1 + "," + bw46_2
return (ag_list, bw46_list, sorted_gf)
def genotype_ags(genotype_list, pop):
""" This functions assigns the most common antigens to a genotype list from locus"""
ag_freq_1 = 0.0
ag_freq_2 = 0.0
geno_antigen_freq = {}
for genotype in genotype_list:
allele_1 = genotype.split("+")[0]
allele_1 = allele_1.rstrip("g p P G")
allele_1 = hla.allele_truncate(allele_1)
allele_2 = genotype.split("+")[1]
allele_2 = allele_2.rstrip("g p P G")
allele_2 = hla.allele_truncate(allele_2)
ag_1 = allele_to_ag_dict[allele_1][0]
bw46_1 = allele_to_ag_dict[allele_1][2]
ag_2 = allele_to_ag_dict[allele_2][0]
bw46_2 = allele_to_ag_dict[allele_2][2]
if allele_1 in population_allele_frequencies[pop]:
ag_freq_1 = population_allele_frequencies[pop][allele_1]
else:
ag_freq_1 = 0.0
if allele_2 in population_allele_frequencies[pop]:
ag_freq_2 = population_allele_frequencies[pop][allele_2]
else:
ag_freq_2 = 0.0
gf = 0
if (ag_1 == ag_2):
gf = float(ag_freq_1) * float(ag_freq_2)
else:
gf = 2 * float(ag_freq_1) * float(ag_freq_2)
geno_antigen = ag_1 + "+" + ag_2
# print (allele_1 + " " + allele_2 + " " + ag_1 + " " + ag_2 + " " + str(ag_freq_1) + " " + str(ag_freq_2) + " " + str(gf))
if geno_antigen in geno_antigen_freq.keys():
geno_antigen_freq[geno_antigen] += float(gf)
else:
geno_antigen_freq[geno_antigen] = float(gf)
TF = sum(geno_antigen_freq.values())
if TF == 0.0:
TF = 1
else:
TF = TF
for i,j in geno_antigen_freq.items():
ag_probs = j/TF
#ag_probs = round(ag_probs, 4)
geno_antigen_freq[i] = ag_probs
#print(geno_antigen_freq)
sorted_gf = sorted(geno_antigen_freq.items(), key = operator.itemgetter(1), reverse = True)
#print(sorted_gf)
#if len(sorted_gf) == 1:
#ag_prob = 1
#else:
#antigen_list = sortef_gf[1::2]
top_ag_geno = sorted_gf[0][0]
top_gf = sorted_gf[0][1]
#print(top_ag_geno)
ag_1 = top_ag_geno.split("+")[0]
ag_2 = top_ag_geno.split("+")[1]
#print(ag_1)
#print(ag_2)
ag_list = ag_1 + "," + ag_2
#print(ag_list)
bw46_list = bw46_1 + "," + bw46_2
#print(bw46_list)
return (ag_list, bw46_list, sorted_gf)