def Translation(seq):
'''translate DNA to protein'''
length = len(seq) / 3
protein = ''
for i in range(length):
if Codon2AA2(seq[(i * 3):((i + 1) * 3)]) == 'J': ## stop codon use *
tmpAA = '*'
elif Codon2AA2(seq[(i * 3):((i + 1) * 3)]) == 'Z': ## IUPAC code
#print seq[(i*3):((i+1)*3)]
tmp_3mer_list = IUPAC_3mer(seq[i * 3:(i + 1) * 3])
tmp_aa_list = []
for tmp_3mer in tmp_3mer_list:
tmp_aa_list.append(Codon2AA2(tmp_3mer))
if len(set(tmp_aa_list)) > 1:
tmpAA = 'X' ## X represents any aa
elif len(set(tmp_aa_list)) == 1:
tmpAA = tmp_aa_list[0]
else:
tmpAA = '*'
else:
tmpAA = Codon2AA2(seq[i * 3:(i + 1) * 3])
protein += tmpAA
return protein
def IUPAC_3mer(seq):
'''Return a list of all possible 3mers of the sequence'''
kmer_list = []
for dna1 in _IUPAC[seq[0]]:
for dna2 in _IUPAC[seq[1]]:
for dna3 in _IUPAC[seq[2]]:
if Codon2AA2(dna1+dna2+dna3) != "J":
kmer_list.append(dna1+dna2+dna3)
return kmer_list
def GetEDP(seq, transcript_len):
'''get features including: ORF length, ORF ratio, ORF EDP of codon'''
# entropy density
Codon = {}
for aa in _AA_list:
Codon[aa] = 1e-9
sum_codon = 1e-9 * 20
if (len(seq) > 3):
num = len(seq) / 3
for i in range(0, num):
if Codon2AA2(seq[i * 3:(i + 1) * 3]) == "J":
continue
# consider the IUPAC codon
elif Codon2AA2(seq[i * 3:(i + 1) * 3]) == "Z":
tmp_kmer_list = IUPAC_3mer(seq[i * 3:(i + 1) * 3])
for tmp_kmer in tmp_kmer_list:
Codon[Codon2AA2(tmp_kmer)] += 1.0 / len(tmp_kmer_list)
sum_codon += 1.0
else:
Codon[Codon2AA2(seq[i * 3:(i + 1) * 3])] += 1.0
sum_codon += 1.0
H = 0.0
for (k, v) in Codon.items():
Codon[k] /= sum_codon
Codon[k] = -Codon[k] * np.log2(Codon[k])
H += Codon[k]
EDP = {}
for (k, v) in Codon.items():
EDP[k] = Codon[k] / H
if EDP[k] < 1e-7:
EDP[k] = 0
outline = ''
for (k, v) in EDP.items():
outline += str(v) + "\t"
return outline.strip()
def SixMer2AA(seq):
'''Convert 6mer to 2 AA'''
return Codon2AA2( seq[0:3] ) + Codon2AA2( seq[3:6] )