from dnachisel import DnaOptimizationProblem, AvoidPattern, random_dna_sequence
from urllib import request
# DOWNLOAD THE LIST OF TF BINDING SITES
url = "http://regulondb.ccg.unam.mx/menu/download/datasets/files/PSSMSet.txt"
data = request.urlopen(url).read().decode('utf-8')
# PARSE THE DATA LINE BY LINE TO OBTAIN A LIST OF TF BINDING SEQUENCES
tf_binding_sequences = [
line for line in data.splitlines() if set() < set(line) <= set("ATGC")
]
# DEFINE AND SOLVE THE OPTIMIZATION PROBLEM
problem = DnaOptimizationProblem(
sequence=random_dna_sequence(50000),
constraints=[AvoidPattern(pattern) for pattern in tf_binding_sequences])
problem.resolve_constraints()
problem.to_record("sequence_without_tf_binding_sites.gb")
naive_construct.features.append(whole_seq_feat)
dest_feat = find_annotation(naive_construct, placeholder.name)
dest_loc = Location.from_biopython_location(dest_feat.location)
user_objectives, user_constraints = load_user_options(args, dest_loc)
objectives += user_objectives
constraints += user_constraints
problem = DnaOptimizationProblem(str(naive_construct.seq), constraints=constraints, objectives=objectives)
domesticator_record = problem.to_record()
mature_construct = naive_construct
mature_construct.features.extend(domesticator_record.features)
SeqIO.write([mature_construct], output_filename, "genbank")
exit("exported " + output_filename)
inserts = load_inserts(args.input)
#now load all the custom global constraints and objectives?
outputs = []
url = (
"https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?"
+ "db=nucleotide&id=48994873&rettype=gb&retmode=txt"
)
genbank_data = request.urlopen(url).read().decode("utf-8")
genbank_record = load_record(StringIO(genbank_data), file_format="genbank")
print("INITIALIZING THE PROBLEM WITH CONSTRAINTS FOR EACH GENE...")
constraints = []
for feature in genbank_record.features:
if feature.type == "gene" and len(feature.location.parts) == 1:
location = Location.from_biopython_location(feature.location)
if (len(location) % 3 == 0) and len(location) > 100:
gene_constraints = [
EnforceTranslation(location = location),
AvoidPattern("BsmBI_site", location),
EnforceGCContent(
mini=0.40, maxi=0.60, window=150, location=location
),
]
constraints.extend(gene_constraints)
problem = DnaOptimizationProblem(genbank_record, constraints)
print("RESOLVING THE CONSTRAINTS...")
problem.logger.ignore_bars_under = 50
problem.resolve_constraints()
problem.to_record("ecoli_genes_optimization.gb")
def domesticate(
self,
dna_sequence=None,
protein_sequence=None,
is_cds="default",
codon_optimization=None,
extra_constraints=(),
extra_objectives=(),
final_record_target=None,
edit=False,
barcode="",
barcode_spacer="AA",
report_target=None,
):
"""Domesticate a sequence.
Parameters
----------
dna_sequence
The DNA sequence string to domesticate.
protein_sequence
Amino-acid sequence of the protein, which will be converted into
a DNA sequence string.
is_cds
If True, sequence edits are restricted to synonymous mutations.
codon_optimization
Either None for no codon optimization or the name of an organism
supported by DnaChisel.
extra_constraints
List of extra constraints to apply to the domesticated sequences.
Each constraint is either a DnaChisel constraint or a function
(dna_sequence => DnaChisel constraint).
extra_objectives
List of extra optimization objectives to apply to the domesticated
sequences. Each objective is either a DnaChisel constraint or a
function (dna_sequence => DnaChisel constraint).
final_record_target
Path to the file where to write the final genbank.
edit
Turn to True to allow sequence edits (if it is false and no all
constraints are originally satisfied, a failed domestication result
(i.e. with attribute ``success`` set to False) will be returned.
report_target
Target for the sequence optimization report (a folder path, or a zip
path).
barcode
A sequence of DNA that will be added to the left of the sequence once
the domestication is done.
barcode_spacer
Nucleotides to be added between the barcode and the enzyme (optional,
the idea here is that they will make sure to avoid the creation of
unwanted cutting sites).
Returns
-------
final_record, edits_record, report_data, success, msg
"""
if is_cds == "default":
is_cds = self.cds_by_default
if isinstance(dna_sequence, SeqRecord):
problem = DnaOptimizationProblem.from_record(dna_sequence)
for spec in problem.constraints + problem.objectives:
spec.location += len(self.left_flank)
extra_constraints = list(extra_constraints) + problem.constraints
extra_objectives = list(extra_constraints) + problem.objectives
if protein_sequence is not None:
is_cds = True
dna_sequence = reverse_translate(protein_sequence)
constraints = [
c(dna_sequence) if hasattr(c, "__call__") else c
for c in list(extra_constraints) + self.constraints
]
location = Location(len(self.left_flank),
len(self.left_flank) + len(dna_sequence))
if is_cds:
constraints.append(EnforceTranslation(location=location))
objectives = [
o(dna_sequence) if hasattr(o, "__call__") else o
for o in list(extra_objectives) + self.objectives
]
if codon_optimization:
objectives.append(
CodonOptimize(species=codon_optimization, location=location))
if self.minimize_edits:
objectives.append(AvoidChanges())
extended_sequence = self.left_flank + dna_sequence + self.right_flank
if (not is_cds) and (not edit):
constraints.append(AvoidChanges())
problem = DnaOptimizationProblem(
extended_sequence,
constraints=constraints,
objectives=objectives,
logger=self.logger,
)
all_constraints_pass = problem.all_constraints_pass()
no_objectives = (len(problem.objectives) - self.minimize_edits) == 0
report_data = None
optimization_successful = True
message = ""
# print (all_constraints_pass, no_objectives)
if not (all_constraints_pass and no_objectives):
problem.n_mutations = self.simultaneous_mutations
if report_target is not None:
(success, message, report_data) = problem.optimize_with_report(
target=report_target, project_name=self.name)
optimization_successful = success
else:
report_data = None
try:
problem.resolve_constraints()
problem.optimize()
except Exception as err:
message = str(err)
optimization_successful = False
report_data = None
final_record = problem.to_record(
with_original_features=True,
with_original_spec_features=False,
with_constraints=False,
with_objectives=False,
)
edits_record = problem.to_record(
with_original_features=True,
with_original_spec_features=False,
with_constraints=False,
with_objectives=False,
with_sequence_edits=True,
)
if final_record_target is not None:
SeqIO.write(final_record, final_record_target, "genbank")
return DomesticationResult(
problem.sequence_before,
final_record,
edits_record,
report_data,
optimization_successful,
message,
)
