def load_template(filename, insert, destination): ''' func descriptor ''' objectives = [] constraints = [] vector = SeqIO.read(filename, "genbank") vector, insert_location = insert_into_vector(vector, destination, insert) problem = DnaOptimizationProblem.from_record(vector) constraints += problem.constraints objectives += problem.objectives #feats = [feat.qualifiers for feat in vector.features] #dnachisel hasn't implemented MultiLocation yet #vector_location = FeatureLocation(insert_location.end, len(vector)) + FeatureLocation(0,insert_location.start) #vector_location_us = Location(0, insert_location.start, 1) #vector_location_ds = Location(insert_location.end, len(vector), 1) #constraints.append(EnforceTranslation(Location.from_biopython_location(insert_location))) #constraints.append(AvoidChanges(vector_location_us)) #constraints.append(AvoidChanges(vector_location_ds)) #This seq should be a SeqRecord object return vector, objectives, constraints
def work(self):
data = self.data
self.logger(message='Initializing...')
if data.editFeatures:
record = sequence_to_biopython_record(data.sequence.upper())
for feature in sorted(data.editedFeatures.values(),
key=lambda f: (f.start, f.end)):
annotate_record(record,
feature_type="misc_feature",
location=(feature.start, feature.end),
label=feature.label)
else:
records, fmt = records_from_data_file(data.file)
record = records[0]
problem = DnaOptimizationProblem.from_record(record)
problem.max_random_iters = 1000
problem.logger = self.logger
success, summary, zip_data = optimization_with_report(
target="@memory", problem=problem, project_name=record.id)
return {
'zip_file': {
'data': data_to_html_data(zip_data, 'zip'),
'name': 'optimization_report.zip',
'mimetype': 'application/zip'
},
'success': success,
'summary': summary
}
def work(self):
data = self.data
self.logger(message="Initializing...")
if data.editFeatures:
record = sequence_to_biopython_record(data.sequence.upper())
for feature in sorted(data.editedFeatures.values(),
key=lambda f: (f.start, f.end)):
annotate_record(
record,
feature_type="misc_feature",
location=(feature.start, feature.end),
label=feature.label,
)
else:
record = records_from_data_files([data.file])[0]
problem = DnaOptimizationProblem.from_record(record,
logger=self.logger)
problem.optimization_stagnation_tolerance = 30
success, summary, zip_data = problem.optimize_with_report(
target="@memory", project_name=record.id)
return {
"zip_file": {
"data": data_to_html_data(zip_data, "zip"),
"name": "optimization_report.zip",
"mimetype": "application/zip",
},
"success": success,
"summary": summary,
}
def test_rca_example():
"""Test a Genbank with ~harmonize_rca feature."""
path = os.path.join("tests", "tests_from_genbanks", "genbanks", "rca_example.gb")
problem = DnaOptimizationProblem.from_record(path)
assert str(problem.objectives) == "[HarmonizeRCA[0-105(+)](e_coli -> h_sapiens)]"
assert problem.objectives[0].original_species == "e_coli"
assert problem.objectives[0].species == "h_sapiens"
problem.optimize()
def test_record_with_multispec_feature():
sequence = random_dna_sequence(100)
record = sequence_to_biopython_record(sequence)
label = "@gc(40-60%/20bp) & @no(BsaI_site) & @keep"
annotate_record(record, label=label)
problem = DnaOptimizationProblem.from_record(record)
assert len(problem.constraints) == 3
c1, c2, c3 = problem.constraints
assert c1.mini == 0.4
assert c2.pattern.name == "BsaI"
def test_cuba_example_1():
path = os.path.join('tests', 'tests_from_genbanks', 'genbanks',
'cuba_example_1.gbk')
record = load_record(path)
problem = DnaOptimizationProblem.from_record(record)
assert not problem.all_constraints_pass()
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert problem.objective_scores_sum() < -100
problem.optimize()
assert problem.objective_scores_sum() > -0.1
def test_cuba_example_1():
path = os.path.join(
"tests", "tests_from_genbanks", "genbanks", "cuba_example_1.gbk"
)
problem = DnaOptimizationProblem.from_record(path)
assert not problem.all_constraints_pass()
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert problem.objective_scores_sum() < -100
problem.optimize()
assert problem.objective_scores_sum() > -0.1
def test_all_shorthands():
"""This test compiles all shorthands as a check that nothing is broken."""
numpy.random.seed(123)
sequence = random_dna_sequence(1000)
record = sequence_to_biopython_record(sequence)
annotate_record(record, (100, 900), label="@no(CATG)")
annotate_record(record, (100, 900), label="@gc(40-60%)")
annotate_record(record, (100, 900), label="@insert(AarI_site)")
annotate_record(record, (650, 752), label="@cds")
annotate_record(record, (100, 200), label="@keep")
annotate_record(record, (250, 273), label="@primer")
annotate_record(record, (250, 280), label="@change")
annotate_record(record, (943, 950), label="@sequence(AKGNTKT)")
annotate_record(record, (955, 958), label="@sequence(ATT|ATC|GGG)")
problem = DnaOptimizationProblem.from_record(record)
assert len(problem.constraints) == 13 # AllowPrimer counts for 4 specs.
assert not problem.all_constraints_pass()
problem.resolve_constraints()
assert problem.all_constraints_pass()
"""Basic demo of the high-level method optimize_with_report."""
import os
from dnachisel import DnaOptimizationProblem
genbank_path = os.path.join("data", "example_sequence.gbk")
report_folder = os.path.join("reports", "optimization_with_report")
problem = DnaOptimizationProblem.from_record(genbank_path)
success, message, _ = problem.optimize_with_report(target=report_folder)
print(message + " A report was generated in " + report_folder)
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,
)
def test_genbank_import_from_filepath():
problem = DnaOptimizationProblem.from_record(example_sequence_path)
assert len(problem.constraints) == 5
assert len(problem.objectives) == 3
def test_genbank_import_from_record_unknown_specs():
record = load_record(example_sequence_path)
with pytest.raises(TypeError):
_ = DnaOptimizationProblem.from_record(
record, specifications_dict={}
)
EnforceTranslation(location=(500, 1400))
],
objectives=[CodonOptimize(species='e_coli', location=(500, 1400))])
# Solve the constraints, optimize with respect to the objective
problem.resolve_constraints()
problem.optimize()
# Print sumarries to check that constraints
# print(problem.constraints_text_summary())
# print(problem.objectives_text_summary())
# ADD NOTES
problemTwo = DnaOptimizationProblem.from_record("my_record.gb")
problemTwo.optimize_with_report(target="report.zip")
'''
===> SUCCESS - all constraints evaluations pass
✔PASS ┍ AvoidPattern[0-10000](pattern:BsaI(GGTCTC))
│ Passed. Pattern not found !
✔PASS ┍ EnforceGCContent[0-10000](mini:0.30, maxi:0.70, window:50)
│ Passed !
✔PASS ┍ EnforceTranslation[500-1400(+)]
│ Enforced by nucleotides restrictions
===> TOTAL OBJECTIVES SCORE: -8.83
-8.83 ┍ MaximizeCAI[500-1400](e_coli)
│ Codon opt. on window 500-1400 scored -8.83E+00
