def calculate_codon_properties(cds_dict, args):
print("\n\nPerforming codon analyses now\n")
pcg = ['cox1','cox3','atp6','atp8','nad4l','nad4','nad6','nad1','nad5','cob','nad2','nad3','cox2']
cds_dict = filter_ambiguous(cds_dict, pcg)
reference = []
i = 1
cont = -1
for tupl in cds_dict.keys():
print(' ' + str(round(i/len(cds_dict.keys())*100, 2)) + '%', end = "\r")
i += 1
for gene in pcg:
if gene in tupl:
cai_value = "NA"
sequence = cds_dict[tupl]
sequence = sequence[:-3]
rscu_list = []
rscu_list.append(sequence)
if args.CAI == True:
for other_tpl in cds_dict.keys():
cont += 1
if other_tpl != tupl:
if list(cds_dict.keys())[cont][4] == args.CAI:
reference.append(cds_dict[other_tpl])
cont = -1
cai_value = CAI(sequence, reference= reference, genetic_code= int(args.GeneCode))
rscu_values = RSCU(rscu_list, genetic_code= int(args.GeneCode))
write_codon_properties(file = tupl, cai = cai_value, rscu = rscu_values, args = args)
def test_cai():
# first, make sure all arguments get the same result
assert CAI("AAC", reference=["AAC"]) == CAI("AAC", RSCUs=RSCU(["AAC"])) == CAI("AAC", weights=relative_adaptiveness(sequences=["AAC"])) == 1.0
# check other sequences
assert CAI("AAT", reference=["AAC"]) == 0.5
assert CAI("AATAAT", reference=["AAC"]) == 0.5
assert CAI("AAT"*100, reference=["AAC"]) == 0.5
def test_bad_args():
# make sure bad arguments raise errors
with pytest.raises(TypeError):
CAI("AAC") # no reference data
with pytest.raises(TypeError):
CAI("AAC", reference=["AAC"], RSCUs=RSCU(["AAC"]))
def test_stop_codon():
# stop codons should be equivalent to an empty string since they don't have RSCUs
assert RSCU(["TAA"]) == RSCU(["TAG"]) == RSCU([" "])
def test_multiple_seqs():
# multiple sequences should be identical to their concatenation
assert RSCU(["AAC", "ATC"]) == RSCU(["AACATC"])
assert RSCU(["AAC", "ATC", "AACGATACGGCACGT"]) == RSCU(["AACATCAACGATACGGCACGT"])
def test_seq():
# make sure module works on Bio.Seq objects
from Bio.Seq import Seq
assert RSCU([Seq("AGC")]) == RSCU(["AGC"])
assert RSCU([Seq("AACGATACGGCACGT")]) == RSCU(["AACGATACGGCACGT"])
def test_str_arg():
# raise an error if given a string
with pytest.raises(ValueError):
RSCU("AAA")
def test_sum():
# The sum of the RSCUs should be equal to the number of codons
assert abs(sum(RSCU(["AAC"]).values()) - len(RSCU(["AAC"]))) < 0.0001
assert abs(sum(RSCU(["AACGATACGGCACGT"]).values()) - len(RSCU(["AAC"]))) < 0.0001
def test_rscu():
assert RSCU(["AAC"]) == {
"AAA": 1.0,
"AAC": 1 / (0.5 * (1 + 0.5)),
"AAG": 1.0,
"AAT": 0.5 / (0.5 * (1 + 0.5)),
"ACA": 1.0,
"ACC": 1.0,
"ACG": 1.0,
"ACT": 1.0,
"AGA": 1.0,
"AGC": 1.0,
"AGG": 1.0,
"AGT": 1.0,
"ATA": 1.0,
"ATC": 1.0,
"ATG": 1.0,
"ATT": 1.0,
"CAA": 1.0,
"CAC": 1.0,
"CAG": 1.0,
"CAT": 1.0,
"CCA": 1.0,
"CCC": 1.0,
"CCG": 1.0,
"CCT": 1.0,
"CGA": 1.0,
"CGC": 1.0,
"CGG": 1.0,
"CGT": 1.0,
"CTA": 1.0,
"CTC": 1.0,
"CTG": 1.0,
"CTT": 1.0,
"GAA": 1.0,
"GAC": 1.0,
"GAG": 1.0,
"GAT": 1.0,
"GCA": 1.0,
"GCC": 1.0,
"GCG": 1.0,
"GCT": 1.0,
"GGA": 1.0,
"GGC": 1.0,
"GGG": 1.0,
"GGT": 1.0,
"GTA": 1.0,
"GTC": 1.0,
"GTG": 1.0,
"GTT": 1.0,
"TAC": 1.0,
"TAT": 1.0,
"TCA": 1.0,
"TCC": 1.0,
"TCG": 1.0,
"TCT": 1.0,
"TGC": 1.0,
"TGG": 1.0,
"TGT": 1.0,
"TTA": 1.0,
"TTC": 1.0,
"TTG": 1.0,
"TTT": 1.0,
}
seq_list = []
counter = 0
n_count = 0
seq_object = SeqIO.parse(input_path, "fasta")
for seqs in seq_object:
seq_id = seqs.id
seq = str(seqs.seq)
seq_list.append(seq)
counter += 1
nn = len(seq)
n_count += nn
print("\n" + str(counter) + " genes imported containing " + str(n_count) + " nucleotides")
print("\nCalculating RSCU for imported genes\n")
try:
RSCU_list = RSCU(seq_list)
except:
print("\nEXCEPTION: RSCU could not be caluclated for imported genes")
print("\nParsing RSCU for codon optimization")
#To do: Create RSCU parser function
for k, v in codon_table_11.items():
for k2, v2 in v.items():
if k2 in RSCU_list:
codon_table_11[k][k2] = RSCU_list[k2]
print("\nOptimizing codons for input gene list")
#Read gene fasta sequence and initiate optimizer
problem = DnaOptimizationProblem(
def test_bad_args():
# make sure bad arguments raise errors
with pytest.raises(TypeError):
relative_adaptiveness()
with pytest.raises(TypeError):
relative_adaptiveness(sequences=["AAC"], RSCUs=RSCU(["AAC"]))
def test_arg_equivalence():
# should be able to take either reference sequences or an RSCU dict
assert relative_adaptiveness(sequences=["AAC"]) == relative_adaptiveness(
RSCUs=RSCU(["AAC"]))
def test_rscu():
assert RSCU(["AAC"]) == {
'AAA': 1.0,
'AAC': 1 / (0.5 * (1 + 0.5)),
'AAG': 1.0,
'AAT': 0.5 / (0.5 * (1 + 0.5)),
'ACA': 1.0,
'ACC': 1.0,
'ACG': 1.0,
'ACT': 1.0,
'AGA': 1.0,
'AGC': 1.0,
'AGG': 1.0,
'AGT': 1.0,
'ATA': 1.0,
'ATC': 1.0,
'ATG': 1.0,
'ATT': 1.0,
'CAA': 1.0,
'CAC': 1.0,
'CAG': 1.0,
'CAT': 1.0,
'CCA': 1.0,
'CCC': 1.0,
'CCG': 1.0,
'CCT': 1.0,
'CGA': 1.0,
'CGC': 1.0,
'CGG': 1.0,
'CGT': 1.0,
'CTA': 1.0,
'CTC': 1.0,
'CTG': 1.0,
'CTT': 1.0,
'GAA': 1.0,
'GAC': 1.0,
'GAG': 1.0,
'GAT': 1.0,
'GCA': 1.0,
'GCC': 1.0,
'GCG': 1.0,
'GCT': 1.0,
'GGA': 1.0,
'GGC': 1.0,
'GGG': 1.0,
'GGT': 1.0,
'GTA': 1.0,
'GTC': 1.0,
'GTG': 1.0,
'GTT': 1.0,
'TAC': 1.0,
'TAT': 1.0,
'TCA': 1.0,
'TCC': 1.0,
'TCG': 1.0,
'TCT': 1.0,
'TGC': 1.0,
'TGG': 1.0,
'TGT': 1.0,
'TTA': 1.0,
'TTC': 1.0,
'TTG': 1.0,
'TTT': 1.0
}
return ss.pearsonr(x, y)[0]
elif "spearman" in corelationFunction:
return ss.spearmanr(x, y, nan_policy="omit")[0]
elif "kendall" in corelationFunction:
return ss.kendalltau(x, y, nan_policy="omit")[0]
geneDict = FSB.findSequenceByID(targetFastaFile, idType="raw")
print("Adult Set:")
seqList = []
for gene in geneDict:
seq = geneDict[gene]
seqList.append(seq)
adultRSCU = RSCU(seqList)
correlationList = []
for gene in geneDict:
seq = geneDict[gene]
rscu = RSCU([seq])
gene_profile = RSCU_To_List(rscu)
correlationList.append(
testCorelation(gene_profile, RSCU_To_List(adultRSCU), "spearman"))
#print(correlationList)
plt.figure()
plt.xlabel("Correlation")
plt.xlim(0, 1)
plt.hist(correlationList, bins=20)
plt.show()
from Bio import SeqIO
from Bio.Seq import Seq
from CAI import RSCU
from matplotlib import pyplot as plt
import numpy as np
for seq_record in SeqIO.parse("KRas.gb", "genbank"):
for f in seq_record.features:
rscu_list = []
if f.type == 'CDS':
feature_seq = f.location.extract(seq_record).seq
coding_result = Seq(str(feature_seq))
# 新建一个列表,列表才能作为输入数据
rscu_list.append(coding_result)
#print(rscu_list)
codon_pre = RSCU(rscu_list)
#print(codon_pre)
# 满足rscu不为空就退出
if len(rscu_list):
break
code = []
value = []
for k in codon_pre:
code.append(k)
values = codon_pre[k]
value.append(values)
plt.switch_backend('Agg')
plt.figure(figsize=(20, 6), dpi=80)
width = 0.8