def write_dna_file(path, dna_sequences, need_log=False):
"""
introduction: Writing DNA sequence set to documents.
:param path: File path.
Type: string
:param dna_sequences: Generated DNA sequences.
Type: one-dimensional list(string)
:param need_log: choose to output log file or not.
"""
m = monitor.Monitor()
try:
with open(path, "w") as file:
if need_log:
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Write DNA sequences to file: " + path)
for row in range(len(dna_sequences)):
if need_log:
m.output(row + 1, len(dna_sequences))
file.write("".join(dna_sequences[row]) + "\n")
return dna_sequences
except IOError:
log.output(
log.ERROR, str(__name__), str(sys._getframe().f_code.co_name),
"The file selection operation was not performed correctly. Please execute the operation again!"
)
def get_yyc_rules(need_log=False):
rules = []
temp_rule1 = ["".join(x) for x in itertools.product("01", repeat=4)]
temp_rule2 = ["".join(x) for x in itertools.product("01", repeat=16)]
m = monitor.Monitor()
if need_log:
# noinspection PyProtectedMember
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Find all the available Yin-Yang rules.")
count, step = 0, 0
for base in ["A", "T", "C", "G"]:
for rule1index in range(len(temp_rule1)):
for rule2index in range(len(temp_rule2)):
rule1 = list(map(int, list(temp_rule1[rule1index])))
rule2 = numpy.array(
list(map(int, list(temp_rule2[rule2index])))).reshape(
4, 4).tolist()
if _check(rule1, rule2):
rules.append(YYCRule(rule1, rule2, base, count))
count += 1
step += 1
if need_log:
m.output(step, len(temp_rule1) * len(temp_rule2) * 4)
return rules
def divide_all(matrix, need_log=False):
"""
introduction: Separate data from indexes in binary strings.
:param matrix: The DNA sequence of len(matrix) rows.
Type: Two-dimensional list(int).
:param need_log: need output log.
:returns index, datas: Obtained data sets and index sets in corresponding locations.
Type: One-dimensional list(int), Two-dimensional list(int).
"""
m = monitor.Monitor()
index_binary_length = int(len(str(bin(len(matrix)))) - 2)
if need_log:
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Divide index and data from binary matrix.")
indexs = []
datas = []
for row in range(len(matrix)):
if need_log:
m.output(row + 1, len(matrix))
index, data = divide(matrix[row], index_binary_length)
indexs.append(index)
datas.append(data)
m.restore()
del matrix, m
return indexs, datas
def connect_all(matrix, need_log=False):
"""
introduction: Integrate index and data from the two-dimensional matrix.
:param matrix: Data from input.
Type: Two-dimensional list(int).
:param need_log:
:return new_matrix: Data for output.
Type: Two-dimensional list(int).
"""
m = monitor.Monitor()
index_binary_length = int(len(str(bin(len(matrix)))) - 2)
if need_log:
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Add index in the binary matrix.")
new_matrix = []
for row in range(len(matrix)):
if need_log:
m.output(row + 1, len(matrix))
new_matrix.append(connect(row, matrix[row], index_binary_length))
m.restore()
del matrix, m
return new_matrix
def _sample_init(self, graph):
'''
we prepare the sampling process
- first we expand its edges to nodes, so eden will be able wo work its magic on it
- then we calculate a score for the graph, to see how much we like it
- we setup the similarity checker stop condition
- possibly we are in a multiprocessing process, and this class instance hasnt been used before,
in this case we need to rebuild the postprocessing function .
'''
self.monitorobject=monitor.Monitor(self.monitor)
self.backtrack=self.maxbacktrack
self.last_graphman = None
graphman=self.preprocessor.transform([graph])[0]
graph = graphman.base_graph()
if self.max_core_size_diff > -1:
self.seed_size = len(graph)
self._score(graphman)
self._sample_notes = ''
self._sample_path_score_set = set()
if self.include_seed==False: # make sure that seed never appears,, may happen if there is nothing happening
self._sample_path_score_set.add(graphman._score)
#print 'sample init:',graphman
#draw.graphlearn_draw(graphman.graph())
return graphman
def sort_order(indexes, data_set, need_log=False):
"""
introduction: Restore data in order of index.
:param indexes: The indexes of data set.
:param data_set: The disordered data set, the locations of this are corresponding to parameter "index".
:param need_log: need output log.
:returns matrix: Binary list in correct order.
Type: Two-dimensional list(int).
"""
m = monitor.Monitor()
if need_log:
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Restore data order according to index.")
# additional information checker
flag_index = 0
if max(indexes) > len(indexes):
while True:
if flag_index + 1 not in indexes:
# index to length
flag_index += 1
break
flag_index += 1
if need_log and flag_index > 0:
log.output(
log.NORMAL, str(__name__), str(sys._getframe().f_code.co_name),
"There are " + str(flag_index) + " required bit segments and " +
str(len(indexes) - flag_index) + " additional bit segments")
# noinspection PyUnusedLocal
if flag_index > 0:
matrix = [[0 for _ in range(len(data_set[0]))]
for _ in range(flag_index)]
else:
matrix = [[0 for _ in range(len(data_set[0]))]
for _ in range(len(indexes))]
for index in range(len(matrix)):
matrix[index] = data_set[indexes.index(index)]
if need_log:
m.output(index + 1, len(matrix))
m.restore()
del indexes, data_set, m
return matrix
def write_all_from_binary(path, matrix, size, need_log=False):
"""
introduction: Writing binary matrix to document.
:param path: File path.
Type: string
:param matrix: A matrix in which each row represents a binary segment that will be used for DNA sequence generation.
Type: two-dimensional list(int)
:param size: This refers to file size, to reduce redundant bits when transferring DNA to binary files.
Type: int
:param need_log: choose to output log file or not.
"""
m = monitor.Monitor()
try:
with open(path, "wb+") as file:
if need_log:
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Write file from binary matrix: " + path)
# Change bit to byte (8 -> 1), and write a file as bytes
bit_index = 0
temp_byte = 0
for row in range(len(matrix)):
if need_log:
m.output(row + 1, len(matrix))
for col in range(len(matrix[0])):
bit_index += 1
temp_byte *= 2
temp_byte += matrix[row][col]
if bit_index == 8:
if size >= 0:
file.write(struct.pack("B", int(temp_byte)))
bit_index = 0
temp_byte = 0
size -= 1
except IOError:
log.output(
log.ERROR, str(__name__), str(sys._getframe().f_code.co_name),
"The file selection operation was not performed correctly. Please execute the operation again!"
)
def read_dna_file(path, need_log=False):
"""
introduction: Reading DNA sequence set from documents.
:param path: File path.
Type: string
:return dna_sequences: A corresponding DNA sequence string in which each row acts as a sequence.
Type: one-dimensional list(string)
:param need_log: need output log.
"""
m = monitor.Monitor()
dna_sequences = []
try:
with open(path, "r") as file:
if need_log:
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Read DNA sequences from file: " + path)
# Read current file by line
lines = file.readlines()
for index in range(len(lines)):
if need_log:
m.output(index + 1, len(lines))
line = lines[index]
dna_sequences.append(
[line[col] for col in range(len(line) - 1)])
return dna_sequences
except IOError:
log.output(
log.ERROR, str(__name__), str(sys._getframe().f_code.co_name),
"The file selection operation was not performed correctly. Please execute the operation again!"
)
def read_binary_from_all(path, segment_length=120, need_log=False):
"""
introduction: Reading binary matrix from document.
:param path: File path.
Type: string
:param segment_length: The binary segment length used for DNA sequence generation.
Considering current DNA synthesis technique limitation,
we usually set 120 as default segment length.
:param need_log: choose to output log file or not.
:return matrix: A matrix in which each row represents a binary segment that will be used for DNA sequence generation.
Type: two-dimensional list(int)
"""
m = monitor.Monitor()
try:
# Open selected file
with open(path, mode="rb") as file:
if need_log:
log.output(log.NORMAL, str(__name__),
str(sys._getframe().f_code.co_name),
"Read binary matrix from file: " + path)
size = os.path.getsize(path)
# Set init storage matrix
matrix = [[0 for _ in range(segment_length)]
for _ in range(math.ceil(size * 8 / segment_length))]
row = 0
col = 0
for byte_index in range(size):
if need_log:
m.output(byte_index + 1, size)
# Read a file as bytes
one_byte = file.read(1)
element = list(
map(
int,
list(
str(bin(struct.unpack(
"B", one_byte)[0]))[2:].zfill(8))))
for bit_index in range(8):
matrix[row][col] = element[bit_index]
col += 1
if col == segment_length:
col = 0
row += 1
if int(len(str(bin(len(matrix)))) - 2) * 7 > segment_length:
if need_log:
log.output(
log.WARN, str(__name__),
str(sys._getframe().f_code.co_name),
"The proportion of index in whole sequence may be high. \n"
"It is recommended to increase the length of output DNA sequences "
"or to divide the file into more segment pools")
return matrix, size
except IOError:
log.output(
log.ERROR, str(__name__), str(sys._getframe().f_code.co_name),
"The file selection operation was not performed correctly. Please execute the operation again!"
)
def __init__(self,
base_reference=None,
current_code_matrix=None,
support_bases=None,
support_spacing=0,
max_ratio=0.8,
search_count=100,
max_homopolymer=math.inf,
max_content=1,
min_free_energy=None):
"""
introduction: The initialization method of YYC.
:param base_reference: Correspondence between base and binary data (RULE 1).
Make sure that Two of the bases are 1 and the other two are 0, so there are only 6 case.
:param current_code_matrix: Conversion rule between base and btis based on support and current base (RULE 2).
Label row is the support base, label col is the current base.
A T C G
A X1 Y1 X2 Y2
T X3 Y3 X4 Y4
C X5 Y5 X6 Y6
G X7 Y7 X8 Y8
Make sure that Xn + Yn = 1 and Xn * Yn = 0, n is in [1, 8].
:param support_bases: Base replenishment before official data.
Make sure that the count of support base must more than support spacing.
Make sure that the number range of each position is {0, 1, 2, 3}, reference base index.
:param support_spacing: Spacing between support base and current base.
When the support base is the front of the current base, the spacing is 0.
:param max_ratio: The max ratio of 0 or 1.
When the (count/length) >= this parameter, we decide that this binary sequence is not good.
:param max_homopolymer: maximum length of homopolymer.
:param max_content: maximum content of C and G, which means GC content is in [1 - max_content, max_content].
:param min_free_energy: the free energy of DNA sequence is lower than required min free energy.
"""
# Set default values for Rules 1 and 2 (RULE 495)
if not base_reference:
base_reference = [0, 1, 0, 1]
if not current_code_matrix:
current_code_matrix = [
[1, 1, 0, 0],
[1, 0, 0, 1],
[1, 1, 0, 0],
[1, 1, 0, 0],
]
if not support_bases:
support_bases = [index_base[0]]
# Assign input data to class variables
self.base_reference = base_reference
self.current_code_matrix = current_code_matrix
self.support_bases = support_bases
self.support_spacing = support_spacing
self.max_ratio = max_ratio
self.search_count = search_count
self.max_homopolymer = max_homopolymer
self.max_content = max_content
self.min_free_energy = min_free_energy
# Detect parameters correctness
self._init_check()
self.file_size = 0
self.monitor = monitor.Monitor()
