def test_ungap(self):
seq = Seq.UnknownSeq(7,
alphabet=Alphabet.Gapped(Alphabet.DNAAlphabet(),
"-"))
self.assertEqual("NNNNNNN", str(seq.ungap("-")))
seq = Seq.UnknownSeq(20,
alphabet=Alphabet.Gapped(Alphabet.DNAAlphabet(),
"-"),
character='-')
self.assertEqual("", seq.ungap("-"))
def test_stops(self):
for nucleotide_seq in [
self.misc_stops,
Seq.Seq(self.misc_stops),
Seq.Seq(self.misc_stops, Alphabet.generic_nucleotide),
Seq.Seq(self.misc_stops, Alphabet.DNAAlphabet()),
Seq.Seq(self.misc_stops, IUPAC.unambiguous_dna),
]:
self.assertEqual("***RR", str(Seq.translate(nucleotide_seq)))
self.assertEqual("***RR",
str(Seq.translate(nucleotide_seq, table=1)))
self.assertEqual("***RR",
str(Seq.translate(nucleotide_seq, table="SGC0")))
self.assertEqual("**W**",
str(Seq.translate(nucleotide_seq, table=2)))
self.assertEqual(
"**WRR",
str(Seq.translate(nucleotide_seq,
table="Yeast Mitochondrial")))
self.assertEqual("**WSS",
str(Seq.translate(nucleotide_seq, table=5)))
self.assertEqual("**WSS",
str(Seq.translate(nucleotide_seq, table=9)))
self.assertEqual(
"**CRR",
str(Seq.translate(nucleotide_seq, table="Euplotid Nuclear")))
self.assertEqual("***RR",
str(Seq.translate(nucleotide_seq, table=11)))
self.assertEqual(
"***RR", str(Seq.translate(nucleotide_seq, table="Bacterial")))
def make_seq(seq_data, seq_name):
seq = SeqRecord(id=seq_name,
name=seq_name,
seq=Seq(seq_data['Sequence'],
alphabet=Alphabet.DNAAlphabet()))
seq.features = [
SeqFeature(FeatureLocation(a['start'], a['end'], a['strand']),
type=a['type'],
qualifiers=OrderedDict({
'label': a['label'],
'note': a['note'],
})) for a in seq_data.get('Annotations', [])
]
if not os.path.exists(seq_name):
os.makedirs(seq_name)
with open('{}/README.md'.format(seq_name), 'w') as f:
f.write(seq_data.get('ReadMe', ''))
with open('{}/.sequence.json'.format(seq_name), 'w') as f:
f.write(json.dumps(seq_data))
SeqIO.write(seq, "{}/sequence.fasta".format(seq_name), "fasta")
SeqIO.write(seq, "{}/sequence.gb".format(seq_name), "genbank")
def extract(self, start_pos, end_pos, make_file=False):
range_set = set(range(start_pos, end_pos))
partial_gb = SeqRecord(
Seq(str(self.gb.seq[start_pos:end_pos]), Alphabet.DNAAlphabet()))
for afeat in self.gb.features:
afeat_range = set(range(afeat.location.start, afeat.location.end))
if len(afeat_range & range_set) > 0:
partial_gb.features.append(afeat)
if make_file == True:
record_handle = open(
partial_gb.id + "_" + str(start_pos) + "_" + str(end_pos), "w")
SeqIO.write(partial_gb, record_handle, "genbank")
return partial_gb
except ValueError:
pass
if not isinstance(s, Seq.Seq):
continue # Only Seq has this method
try:
print s.translate()
assert False, "Translation shouldn't work on a protein!"
except ValueError:
pass
misc_stops = "TAATAGTGAAGAAGG"
for nucleotide_seq in [
misc_stops,
Seq.Seq(misc_stops),
Seq.Seq(misc_stops, Alphabet.generic_nucleotide),
Seq.Seq(misc_stops, Alphabet.DNAAlphabet()),
Seq.Seq(misc_stops, IUPAC.unambiguous_dna)
]:
assert "***RR" == str(Seq.translate(nucleotide_seq))
assert "***RR" == str(Seq.translate(nucleotide_seq, table=1))
assert "***RR" == str(Seq.translate(nucleotide_seq, table="SGC0"))
assert "**W**" == str(Seq.translate(nucleotide_seq, table=2))
assert "**WRR" == str(
Seq.translate(nucleotide_seq, table='Yeast Mitochondrial'))
assert "**WSS" == str(Seq.translate(nucleotide_seq, table=5))
assert "**WSS" == str(Seq.translate(nucleotide_seq, table=9))
assert "**CRR" == str(
Seq.translate(nucleotide_seq, table='Euplotid Nuclear'))
assert "***RR" == str(Seq.translate(nucleotide_seq, table=11))
assert "***RR" == str(Seq.translate(nucleotide_seq, table='Bacterial'))
del misc_stops
def gen_components(self):
""" Construct knowledge base components """
# get options
options = self.options
num_chromosomes = options.get('num_chromosomes')
chromosome_topology = options.get('chromosome_topology')
mean_gc_frac = options.get('mean_gc_frac')
mean_num_genes = options.get('mean_num_genes')
mean_gene_len = options.get('mean_gene_len')
mean_coding_frac = options.get('mean_coding_frac')
# generate chromosomes and genes
cell = self.knowledge_base.cell
for i_chr in range(num_chromosomes):
num_genes = self.rand(mean_num_genes / num_chromosomes)[0]
gene_lens = self.rand(mean_gene_len, count=num_genes)
intergene_lens = self.rand(mean_gene_len / mean_coding_frac *
(1 - mean_coding_frac),
count=num_genes)
seq_len = numpy.sum(gene_lens) + numpy.sum(intergene_lens)
seq = Seq.Seq(
''.join(
random.choice(('A', 'C', 'G', 'T'),
p=((1 - mean_gc_frac) / 2, mean_gc_frac / 2,
mean_gc_frac / 2, (1 - mean_gc_frac) / 2),
size=(seq_len, ))), Alphabet.DNAAlphabet())
chr = cell.species_types.get_or_create(
id='chr_{}'.format(i_chr + 1),
__type=wc_kb.core.DnaSpeciesType)
chr.name = 'Chromosome {}'.format(i_chr + 1)
chr.circular = chromosome_topology == 'circular'
chr.double_stranded = True
chr.seq = seq
gene_starts = numpy.int64(
numpy.cumsum(
numpy.concatenate(([0], gene_lens[0:-1])) +
numpy.concatenate((numpy.round(intergene_lens[0:1] / 2),
intergene_lens[1:]))))
for i_gene in range(num_genes):
tu = cell.loci.get_or_create(
id='tu_{}_{}'.format(i_chr + 1, i_gene + 1),
__type=wc_kb.prokaryote.TranscriptionUnitLocus)
tu.polymer = chr
tu.name = 'Transcription unit {}-{}'.format(
i_chr + 1, i_gene + 1)
tu.start = gene_starts[i_gene]
tu.end = gene_starts[i_gene] + gene_lens[i_gene] - 1
tu.strand = random.choice((wc_kb.core.PolymerStrand.positive,
wc_kb.core.PolymerStrand.negative))
gene = cell.loci.get_or_create(
id='gene_{}_{}'.format(i_chr + 1, i_gene + 1),
__type=wc_kb.prokaryote.GeneLocus)
gene.polymer = chr
gene.transcription_units.append(tu)
gene.name = 'Gene {}-{}'.format(i_chr + 1, i_gene + 1)
gene.start = gene_starts[i_gene]
gene.end = gene_starts[i_gene] + gene_lens[i_gene] - 1
gene.type = wc_kb.core.GeneType.mRna
gene.strand = tu.strand
def translate(seq):
return Seq(seq.replace('-', 'N'),
Alphabet.DNAAlphabet()).translate().tostring()
