def test_EnforceTranslation_bacterial_valine_antisense():
table_name = "Bacterial"
protein = "LLTMMVTTTTVMVL"
protein_sequence = reverse_translate(protein, table=table_name)
for first_codon_before, first_codon_after in [
("ATG", "ATG"), # methionine stays the only methionine codon
("GTG", "GTG"), # valine-start-codon stays the only valine-start-codon
]:
sequence = first_codon_before + protein_sequence
cds_constraint = EnforceTranslation(
genetic_table="Bacterial",
start_codon="keep",
location=Location(0, len(sequence), -1),
)
problem = DnaOptimizationProblem(
sequence=reverse_complement(sequence),
constraints=[cds_constraint],
objectives=[EnforceChanges()],
logger=None,
)
assert problem.constraints[0].translation == "MLLTMMVTTTTVMVL"
problem.optimize()
problem_sequence_rv = reverse_complement(problem.sequence)
protein_after = translate(
problem_sequence_rv, table_name, assume_start_codon=True
)
assert protein_after == "M" + protein
assert problem_sequence_rv[:3] == first_codon_after
def test_AvoidRareCodons_as_constraint_reversed():
numpy.random.seed(123)
sequence = "ATG" "TTT" "ATA" "CCA" "CTT" "TAG"
rev_sequence = reverse_complement(sequence)
location = (0, len(sequence), -1)
problem = DnaOptimizationProblem(
sequence=rev_sequence,
constraints=[
EnforceTranslation(location=location),
AvoidRareCodons(0.11, "e_coli", location=location),
],
)
assert problem.all_constraints_pass()
assert problem.sequence_edits_as_array().sum() == 4
new_sequence = reverse_complement(problem.sequence)
assert translate(new_sequence) == translate(sequence)
def test_EnforceTranlationReversed():
numpy.random.seed(1234)
sequence = reverse_translate(random_protein_sequence(50, seed=123))
rev_sequence = reverse_complement(sequence)
problem = DnaOptimizationProblem(
sequence=rev_sequence,
constraints=[
AvoidPattern("AGC"),
EnforceTranslation(location=(0, len(sequence), -1))
],
)
assert not problem.all_constraints_pass()
problem.resolve_constraints()
assert problem.all_constraints_pass()
def test_AvoidRareCodons_as_objective_reversed():
numpy.random.seed(123)
sequence = "ATG" "TTT" "ATA" "CCA" "CTT" "TAG"
rev_sequence = reverse_complement(sequence)
location = (0, len(sequence), -1)
problem = DnaOptimizationProblem(
sequence=rev_sequence,
constraints=[EnforceTranslation(location=location)],
objectives=[AvoidRareCodons(0.11, "e_coli", location=location)],
)
assert abs(problem.objective_scores_sum() + 0.09) < 0.001
problem.optimize()
assert problem.objective_scores_sum() == 0
def test_avoid_hairpin_basics():
numpy.random.seed(123)
random_sequences = [random_dna_sequence(30) for i in range(10)]
full_sequence = "".join([
seq
for sequence in random_sequences
for seq in (random_dna_sequence(50),
sequence,
random_dna_sequence(50),
reverse_complement(sequence),
random_dna_sequence(50))
])
problem = DnaOptimizationProblem(full_sequence,
constraints=[AvoidHairpins()])
assert not problem.all_constraints_pass()
problem.resolve_constraints()
assert problem.all_constraints_pass()
"""Example of use of the AvoidPAttern specification"""
from dnachisel import (DnaOptimizationProblem, random_dna_sequence,
reverse_complement, AvoidHairpins)
random_sequences = [random_dna_sequence(30) for i in range(10)]
full_sequence = "".join([
seq for sequence in random_sequences
for seq in (random_dna_sequence(50), sequence, random_dna_sequence(50),
reverse_complement(sequence), random_dna_sequence(50))
])
problem = DnaOptimizationProblem(full_sequence, constraints=[AvoidHairpins()])
print("\nBefore optimization:\n")
print(problem.constraints_text_summary())
problem.resolve_constraints()
print("\nAfter optimization:\n")
print(problem.constraints_text_summary())
def test_enforce_pattern_options():
# Checks for Github issue #53
# Test 6 cases: location yes/no, 3 strand options
sequence = "A" * 10
pattern = "C" * 4
# location=None
problem = dc.DnaOptimizationProblem(
sequence=sequence,
constraints=[
dc.EnforcePatternOccurence(pattern,
occurences=1,
strand="from_location"),
],
logger=None,
)
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert pattern in problem.sequence
problem = dc.DnaOptimizationProblem(
sequence=sequence,
constraints=[
dc.EnforcePatternOccurence(pattern, occurences=1, strand="both")
],
logger=None,
)
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert pattern in problem.sequence
problem = dc.DnaOptimizationProblem(
sequence=sequence,
constraints=[
dc.EnforcePatternOccurence(pattern, occurences=1, strand=-1)
],
logger=None,
)
assert problem.constraints[0].evaluate(problem).score == -1
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert dc.reverse_complement(
pattern) in problem.sequence # other strand used
# location specificed
# Use -1 strand from location:
problem = dc.DnaOptimizationProblem(
sequence=sequence,
constraints=[
dc.EnforcePatternOccurence(
pattern,
occurences=1,
strand="from_location",
location=Location(1, 6, strand=-1),
)
],
logger=None,
)
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert dc.reverse_complement(pattern) in problem.sequence
# Overwrite -1 strand to "both":
problem = dc.DnaOptimizationProblem(
sequence=sequence,
constraints=[
dc.EnforcePatternOccurence(
pattern,
occurences=1,
strand="both",
location=Location(1, 6, strand=-1),
)
],
logger=None,
)
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert pattern in problem.sequence # uses +1 strand by default
# Overwrite -1 strand to +1:
problem = dc.DnaOptimizationProblem(
sequence=sequence,
constraints=[
dc.EnforcePatternOccurence(
pattern,
occurences=1,
strand=1,
location=Location(1, 6, strand=-1),
)
],
logger=None,
)
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert pattern in problem.sequence # uses +1 strand
