def test_that_constraints_in_records_are_accounted_for():
sequence = "ATACGTCTCTAG"
rec = sequence_to_biopython_record(sequence)
annotate_record(rec, label="@cds")
from genedom import BUILTIN_STANDARDS
emma = BUILTIN_STANDARDS["EMMA"]
p7 = emma.domesticators["p7"]
result = p7.domesticate(rec)
seq_after = str(
result.record_after[len(p7.left_flank):-len(p7.right_flank)].seq)
assert translate(seq_after) == translate(sequence)
def test_AvoidRareCodons_as_constraint():
numpy.random.seed(123)
sequence = "ATG" "TTT" "ATA" "CCA" "CTT" "TAG"
problem = DnaOptimizationProblem(
sequence=sequence,
constraints=[EnforceTranslation(),
AvoidRareCodons(0.11, "e_coli")],
)
assert problem.all_constraints_pass()
assert problem.sequence_edits_as_array().sum() == 4
assert translate(problem.sequence) == translate(sequence)
def test_EnforceTranslation_bacterial_valine():
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"
)
problem = DnaOptimizationProblem(
sequence=sequence,
constraints=[cds_constraint],
objectives=[EnforceChanges()],
logger=None,
)
assert problem.constraints[0].translation == "MLLTMMVTTTTVMVL"
problem.optimize()
protein_after = translate(
problem.sequence, table_name, assume_start_codon=True
)
assert protein_after == "M" + protein
assert problem.sequence[: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_AvoidStopCodons():
numpy.random.seed(123)
problem = DnaOptimizationProblem(
sequence="".join(["ATT", "TAG", "GCC", "TGA", "ATC", "TAA"]),
constraints=[AvoidStopCodons()],
logger=None,
)
assert not problem.all_constraints_pass()
problem.resolve_constraints()
assert problem.all_constraints_pass()
assert "*" not in translate(problem.sequence)
def test_maximal_protein_sequence_change():
np.random.seed(123)
protein = dc.random_protein_sequence(200)
sequence = dc.reverse_translate(protein)
problem = dc.DnaOptimizationProblem(
sequence=sequence,
constraints=[dc.EnforceTranslation()],
objectives=[dc.EnforceChanges()],
)
problem.resolve_constraints()
problem.optimize()
assert problem.number_of_edits() == 238
assert dc.translate(problem.sequence) == protein
