def gui_CountMotif(self):
pattern = ''
countMotifsList = []
HTMLFILE = 'E:\MY CODES\PYTHON\DNA-Toolkit\Html-Output\motifCount.html'
f = open(HTMLFILE, 'w')
t = html.Table(header_row=self.motifs)
# Length of motifs
for i in range(len(self.sequences)):
dna = self.sequences[i]
my_dna = Seq(dna, generic_dna)
countMotifs = []
countMotifs.append('<b>' + str(i + 1) + '</b>') # number of sequence in table
for j in range(1, len(self.motifs)):
pattern = self.motifs[j]
# Number of repeat pattern
countMotifs.append(my_dna.count_overlap(pattern))
t.rows.append(countMotifs)
htmlcode = str(t)
# print (htmlcode)
f.write(htmlcode)
f.write('<p>')
#print('Done' + '-' * 79)
webbrowser.open(HTMLFILE)
def get_features_for_sequence(dna_seq):
"""
For the given sequence, it returns as features the frequencies of occurrence (as percentages) for all combinations of 1, 2 and 3 consecutive
nucleotides (letters).
E.g. for sequence ACGGT, we have:
- A: 1/5 = 0.2; C: 1/5 = 0.2; G: 2/5 = 0.4; T: 1/5 = 0.2 (we divide by 5 because there are 5 letters in the sequence)
- AA: 0; AC: 1/4 = 0.25; AG: 0; AT: 0; CC: 0; CG = 1/4 = 0.25, etc (we divide by 4 because there are 4 overlapping combinations of 2 letters)
- AAA: 0; AAC: 0; ... ACG: 1/3 = 0.33; ... CGG: 1/3, etc (we divide by 3 because there are 3 overlapping combinations of 3 letters)
The feature map will be: ['A': 0.2, 'C': 0.2, 'G': 0.4, 'T': 0.2, 'AA': 0, 'AC': 0.25, 'AG': 0, 'AT': 0, ..., 'TTT': 0]
"""
seq_list = generate_possible_sequences(3)
seq = Seq(dna_seq)
feature_map = {}
for combination in seq_list:
# We will count the overlapping values and divide them according to their length
divide_number = 1
if len(combination) == 1:
divide_number = len(dna_seq)
elif len(combination) == 2:
divide_number = len(dna_seq) - 1
elif len(combination) == 3:
divide_number = len(dna_seq) - 2
feature_map[combination] = seq.count_overlap(
combination) / divide_number
return feature_map
def gui_CountMotif(self):
#Import html.py library for create table
import lib.HTML as html
pattern = ''
countMotifsList = []
HTMLFILE = 'htmlOutput/motifCount.html'
f = open(HTMLFILE, 'w')
t = html.Table(header_row=self.motifs)
#Length of motifs
for i in range(len(self.sequences)):
dna = self.sequences[i]
my_dna = Seq(dna, generic_dna)
countMotifs = []
countMotifs.append('<b>' + str(i + 1) +
'</b>') #number of sequence in table
for j in range(1, len(self.motifs)):
pattern = self.motifs[j]
#Number of repeat pattern
countMotifs.append(my_dna.count_overlap(pattern))
t.rows.append(countMotifs)
htmlcode = str(t)
f.write(htmlcode)
f.write('<p>')
print('Done' + '-' * 79)
def get_tf_idf_for_sequence(dna_seq, idf_map, size):
"""
:param dna_seq: the current sequence
:param idf_map: a dictionary which was previously populated, containing the IDF for each subsequence (combination: AA, AC, ... TTT),
computed for the entire corpus
:param size: the total number of sequences
:return:
"""
seq_list = ExtractFeatureStrategy.generate_possible_sequences(3)
# remove single letters from the list (they do not bring relevant information in the context of tf-idf, as they occur in all sequences)
seq_list = seq_list[4:]
seq = Seq(dna_seq)
feature_map = {}
for combination in seq_list:
# We will count the overlapping values and divide them according to their length
divide_number = 1
if len(combination) == 2:
divide_number = len(dna_seq) - 1
elif len(combination) == 3:
divide_number = len(dna_seq) - 2
tf = seq.count_overlap(combination) / divide_number
idf = idf_map[combination]
if idf != 0:
idf = math.log2(size / idf)
feature_map[combination] = tf * idf
return feature_map
def count_sequences_containing_subsequence(subsequence, sequences):
"""
From a list of sequences (e.g. an entire corpus), it counts how many times the subsequence occurs.
"""
result = 0
for s in sequences:
seq = Seq(s)
count = seq.count_overlap(subsequence)
if count > 0:
result += 1
return result
def terminal_CountMotif(self):
pattern = ''
countMotifs = {}
# Print length of motifs
for i in range(len(self.sequences)):
dna = self.sequences[i]
my_dna = Seq(dna, generic_dna)
for j in range(1, len(self.motifs)):
pattern = self.motifs[j]
# Number of repeat pattern
countMotifs[pattern] = []
countMotifs[pattern].append(my_dna.count_overlap(pattern))
print(countMotifs)
def terminal_FrequencyMotifs(self):
pattern = ''
countMotifs = {}
for i in range(len(self.sequences)):
dna = self.sequences[i]
my_dna = Seq(dna, generic_dna)
print('Sequence', i + 1)
for j in range(1, len(self.motifs)):
pattern = self.motifs[j]
countMotifs[pattern] = []
countMotifs[pattern].append(my_dna.count_overlap(pattern))
for key, value in countMotifs.items():
if (len(key) == 1):
temp = value[0] #This is a list
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][0]) * 100
) #Caculate Percentage and show 2 decimal
print('Frequency of', key, '==>', percentage, '%')
elif (len(key) == 2):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][1]) * 100)
print('Frequency of', key, '==>', percentage, '%')
elif (len(key) == 3):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][2]) * 100)
print('Frequency of', key, '==>', percentage, '%')
elif (len(key) == 4):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][3]) * 100)
print('Frequency of', key, '==>', percentage, '%')
elif (len(key) == 5):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][4]) * 100)
print('Frequency of', key, '==>', percentage, '%')
def numNmersMotifs(self):
pattern = ''
countMotifs = {}
for i in range(len(self.sequences)):
dna = self.sequences[i]
my_dna = Seq(dna, generic_dna)
summ = [0] * 5
#Define summ variable
for j in range(1, len(self.motifs)):
pattern = self.motifs[j]
countMotifs[pattern] = []
countMotifs[pattern].append(my_dna.count_overlap(pattern))
for value in countMotifs.keys():
if (len(value) == 1):
temp = countMotifs[value] #This is a list
summ[0] += temp[0]
elif (len(value) == 2):
temp = countMotifs[value]
summ[1] += temp[0]
elif (len(value) == 3):
temp = countMotifs[value]
summ[2] += temp[0]
elif (len(value) == 4):
temp = countMotifs[value]
summ[3] += temp[0]
elif (len(value) == 5):
temp = countMotifs[value]
summ[4] += temp[0]
self.numNmersMotifsDic[i] = []
self.numNmersMotifsDic[i].append(summ[0])
self.numNmersMotifsDic[i].append(summ[1])
self.numNmersMotifsDic[i].append(summ[2])
self.numNmersMotifsDic[i].append(summ[3])
self.numNmersMotifsDic[i].append(summ[4])
def kmerFreq(isoform):
K = args.repeat
sline = isoform.rstrip().split()
chrom = sline[0]
start = sline[1]
end = sline[2]
exonCnt = int(sline[9])
exonlen = sline[10].rstrip(',').split(',')
exonlen = [int(length) for length in exonlen]
exonS = sline[11].rstrip(',').split(',')
exonS = [int(s) for s in exonS]
strand = sline[5]
gene_seq = records[chrom].seq[int(sline[1]):int(sline[2])].upper()
gene_seq_str = str(gene_seq)
spliced_seq = ""
for i in range(exonCnt):
spliced_seq = spliced_seq + gene_seq_str[exonS[i]:(exonS[i] + exonlen[i])]
if strand == '-':
gene_seq = gene_seq.reverse_complement()
spliced_seq = Seq(spliced_seq).reverse_complement()
else:
spliced_seq = Seq(spliced_seq)
spliced_length = len(spliced_seq)
gene_length = len(gene_seq)
kmer_freq = [0.0] * (4 ** K)
iso = sline[3]
for kmer in kmer_dict:
if args.overlap:
kmer_freq[kmer_dict[kmer]] = spliced_seq.count_overlap(kmer) + 0.0
else:
kmer_freq[kmer_dict[kmer]] = spliced_seq.count(kmer) + 0.0
if not(args.count):
for ind,cnt in enumerate(kmer_freq):
kmer_freq[ind] = cnt / spliced_length * 1000
kmer_freq = [str(freq) for freq in kmer_freq]
return iso+'\t'+'\t'.join(kmer_freq)+'\n'
def makeDataset():
with open('dataset_disp.csv', 'w') as out:
out.write('seq,ie,gc,nstart,maxnstop,maxdisp\n')
with open('exons_final.txt') as f:
for line in f:
# I/E label
out.write(line[:-1] + ',1,')
# GC
my_seq = Seq(line[:-1], IUPAC.unambiguous_dna)
out.write(str(GC(my_seq)) + ',')
# N_ATG (start codon)
out.write(str(my_seq.count_overlap('ATG')) + ',')
# max(N_TAA, N_TAG, N_TGA) (stop codons)
out.write(
str(
max(my_seq.count_overlap('TAA'),
my_seq.count_overlap('TAG'),
my_seq.count_overlap('TGA'))) + ',')
# max disparity in translation of default frame
#out.write( str(max(disp(trans(my_seq), 15))) + '\n')
with open('introns_final.txt') as f:
for line in f:
# I/E label
out.write(line[:-1] + ',0,')
# GC
my_seq = Seq(line, IUPAC.unambiguous_dna)
out.write(str(GC(my_seq)) + ',')
# N_ATG (start codon)
out.write(str(my_seq.count_overlap('ATG')) + ',')
# max(N_TAA, N_TAG, N_TGA) (stop codons)
out.write(
str(
max(my_seq.count_overlap('TAA'),
my_seq.count_overlap('TAG'),
my_seq.count_overlap('TGA'))) + ',')
def gui_FrequencyMotifs(self):
pattern = ''
countMotifs = {}
percentage = ''
HTMLFILE = 'E:\MY CODES\PYTHON\DNA-Toolkit\Html-Output\motifFrequency.html'
f = open(HTMLFILE, 'w')
t = html.Table(header_row=self.motifs)
for i in range(len(self.sequences)):
dna = self.sequences[i]
my_dna = Seq(dna, generic_dna)
frequencyMotifsList = []
frequencyMotifsList.append('<b>' + str(i + 1) + '</b>') # number of sequence in table
for j in range(1, len(self.motifs)):
pattern = self.motifs[j]
countMotifs[pattern] = []
countMotifs[pattern].append(my_dna.count_overlap(pattern))
for key, value in countMotifs.items():
if (len(key) == 1):
temp = value[0] # This is a list
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][0]) * 100) # Caculate Percentage and show 2 decimal
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 2):
temp = value[0]
percentage = '%.2f' % ((temp / self.numNmersMotifsDic[i][1]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 3):
temp = value[0]
percentage = '%.2f' % ((temp / self.numNmersMotifsDic[i][2]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 4):
temp = value[0]
percentage = '%.2f' % ((temp / self.numNmersMotifsDic[i][3]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 5):
temp = value[0]
percentage = '%.2f' % ((temp / self.numNmersMotifsDic[i][4]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
t.rows.append(frequencyMotifsList)
htmlcode = str(t)
# print (htmlcode)
f.write(htmlcode)
f.write('<p>')
#print('Done' + '-' * 79)
webbrowser.open(HTMLFILE)
from Bio.Seq import Seq
with open("data/rosalind_ini.txt", "r") as sequence_file:
sequence = Seq(sequence_file.read())
#Using count_overlap bcause python string's .count() is a non-overlapping count
#In some biological situations, overlapping count is necessary
print(sequence.count_overlap("A"), sequence.count_overlap("C"),
sequence.count_overlap("G"), sequence.count_overlap("T"))
def gui_FrequencyMotifs(self):
#Import html.py library for create table
import lib.HTML as html
pattern = ''
countMotifs = {}
percentage = ''
HTMLFILE = 'htmlOutput/motifFrequency.html'
f = open(HTMLFILE, 'w')
t = html.Table(header_row=self.motifs)
for i in range(len(self.sequences)):
dna = self.sequences[i]
my_dna = Seq(dna, generic_dna)
frequencyMotifsList = []
frequencyMotifsList.append('<b>' + str(i + 1) +
'</b>') #number of sequence in table
for j in range(1, len(self.motifs)):
pattern = self.motifs[j]
countMotifs[pattern] = []
countMotifs[pattern].append(my_dna.count_overlap(pattern))
for key, value in countMotifs.items():
if (len(key) == 1):
temp = value[0] #This is a list
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][0]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 2):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][1]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 3):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][2]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 4):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][3]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
elif (len(key) == 5):
temp = value[0]
percentage = '%.2f' % (
(temp / self.numNmersMotifsDic[i][4]) * 100)
percentage = str(percentage) + '%'
frequencyMotifsList.append(percentage)
t.rows.append(frequencyMotifsList)
htmlcode = str(t)
f.write(htmlcode)
f.write('<p>')
print('Done' + '-' * 79)
