def revtrans(
input, output, codon_table, codon_usage, sampler, codon_freq_threshold, amber_only
):
"""Reverse translate amino acid sequences into DNA
This operation randomly samples codons for each amino acid, so multiple runs
of this tool on the same input can produce different results
Note: only the Standard codon table is currently implemented.
Note: only the E. coli codon usage is currently implemented.
INPUT and OUTPUT are paths to fasta files or "-" to specify STDIN/STDOUT.
"""
if sampler == "weighted":
usage = ecoli_codon_usage
if codon_freq_threshold is not None:
# TODO: this is hardcoded in and there's a leaky abstraction here
table = standard_dna_table
usage = zero_low_freq_codons(usage, table, codon_freq_threshold)
if amber_only:
usage = zero_non_amber_stops(usage)
codon_sampler = FreqWeightedCodonSampler(usage=usage)
elif sampler == "uniform":
codon_sampler = UniformCodonSampler()
for seqrecord in tqdm(SeqIO.parse(input, "fasta"), desc="revtrans", unit="seq"):
dna_id = seqrecord.id
dna_seq = reverse_translate(seqrecord.seq, codon_sampler)
print_fasta(SeqRecord(dna_seq, dna_id, description=""), output)
def test_freq_weighted_sampler(self):
with warnings.catch_warnings(): # biopython Seq.__hash__
warnings.simplefilter("ignore")
codon_sampler = FreqWeightedCodonSampler(usage=ecoli_codon_usage)
dna_seq = reverse_translate(all_aa_protein_seq, codon_sampler)
assert dna_seq.translate(
table=codon_sampler.table) == all_aa_protein_seq
def test_uniform_sampler(self):
codon_sampler = UniformCodonSampler()
dna_seq = reverse_translate(all_aa_protein_seq, codon_sampler)
assert dna_seq.translate(
table=codon_sampler.table) == all_aa_protein_seq
def test_freq_weighted_sampler(self):
codon_sampler = FreqWeightedCodonSampler(usage=ecoli_codon_usage)
dna_seq = reverse_translate(all_aa_protein_seq, codon_sampler)
assert dna_seq.translate(
table=codon_sampler.table) == all_aa_protein_seq