def test_Transcoder(self):
dna = sequence.DNA(name='test', seq='AAAGTGAGCCCC')
dna.ftable.add_feature(Feature(dna, 1, 3, '+', 'GEN'))
dna.ftable.add_feature(Feature(dna, 4, 9, '+', 'DON'))
dna.ftable.add_feature(Feature(dna, 10, 12, '+', 'GEN'))
tc = hmm.Transcoder(model=self.hmm, dna=dna)
self.assertAlmostEqual(tc.score, 1.250948871669515)
def test_StochasticViterbi_and_Parse(self):
dna = sequence.DNA(name='test', seq='TTTTTGTAAGTAAGTTTTT')
v = hmm.StochasticViterbi(model=self.hmm, dna=dna, seed=1)
ps = v.generate_paths(1000)
self.assertIsInstance(ps[0], hmm.Parse)
self.assertAlmostEqual(ps[0].score, 4.600333948735676)
self.assertAlmostEqual(ps[0].freq, 0.459)
def test_Viterbi_logspace(self):
dna = sequence.DNA(name='test', seq='AAAAGTAAGTTTTT')
m = copy.deepcopy(self.hmm)
m.convert2log()
v = hmm.Viterbi(model=m, dna=dna)
p = v.generate_path()
self.assertAlmostEqual(p.score, 1.5763805776787159)
def test_StochasticViterbi_logspace(self):
dna = sequence.DNA(name='test', seq='TTTTTGTAAGTAAGTTTTT')
m = copy.deepcopy(self.hmm)
m.convert2log()
v = hmm.StochasticViterbi(model=m, dna=dna, seed=1)
ps = v.generate_paths(1000)
self.assertAlmostEqual(ps[0].score, 1.5261288984112191)
self.assertAlmostEqual(ps[0].freq, 0.459)
def test_Viterbi_and_Parse(self):
dna = sequence.DNA(name='test', seq='AAAAGTAAGTTTTT')
v = hmm.Viterbi(model=self.hmm, dna=dna)
p = v.generate_path()
self.assertIsInstance(p, hmm.Parse)
self.assertEqual(p.freq, None)
self.assertEqual(p.path[0], 'GEN')
self.assertAlmostEqual(p.score, 4.837415437963225)
ft = p.features()
self.assertEqual(ft[1].type, 'DON')
self.assertEqual(ft[1].beg, 5)
def test_DNA(self):
s = sequence.DNA(seq='CATGCCAACAATGCNCAY')
self.assertIsInstance(s, sequence.DNA)
try:
s.check_alphabet()
except sequence.SequenceError:
self.fail('DNA object failed check_alphabet()')
p = s.translate()
self.assertIsInstance(p, sequence.Protein)
self.assertEqual(p.seq, 'HANNAH')
s.revcomp()
self.assertIsInstance(s, sequence.DNA)
self.assertEqual(s.seq, 'RTGNGCATTGTTGGCATG')
s2 = sequence.DNA(seq='CATGCCAACAATGCNCAQ')
with self.assertRaises(sequence.SequenceError):
s2.check_alphabet()
with self.assertRaises(sequence.SequenceError):
p2 = s2.translate()
with self.assertRaises(sequence.SequenceError):
s2.revcomp()
def test_ForwardBackward(self):
dna = sequence.DNA(name='test', seq='AGT')
p = hmm.ForwardBackward(model=self.hmm, dna=dna)
self.assertAlmostEqual(p.posterior(state_name='GEN', i=0),
0.021213450734526202,
places=8)
self.assertAlmostEqual(p.posterior(state_name='DON-1', i=2),
0.0031969999999999998,
places=8)
self.assertAlmostEqual(p.posterior(state_name='DON-0', i=1),
0.0031969999999999998,
places=8)
self.assertAlmostEqual(p.posterior(state_name='GEN', i=2),
0.0180164507345262,
places=8)
arg.gff3 + ' ' + '--model ' + 'mRNA2' + ' ' + '--hmm ' +
arg.mRNA2 + ' ' + '--source ' + arg.work +
'C.elegans.mRNA2.source' + ' ' + '--u5_ctx ' + '5 ' +
'--u3_ctx ' + '5 ' + '--koz_ctx ' + '1 ' + '--atg_ctx ' + '2 ' +
'--stop_ctx ' + '2 ' + '--cds_ctx ' + '2')
print('Forge Completed')
fasta = toolbox.FASTA_stream(filename=arg.work +
'C.elegans.mRNA1.source.fasta')
gff = toolbox.GFF_file(filename=arg.work + 'C.elegans.mRNA1.source.gff')
#Extract from GFF3
GFF3_genes = {}
dnas = []
for entry in fasta:
dna = sequence.DNA(seq=entry.seq, name=entry.id)
dnas.append(dna)
gffs = gff.get(chrom=dna.name)
for g in gffs:
if g.type == 'ATG':
GFF3_genes.update({entry.id: g.beg})
#mRNA1 model
mRNA1_genes = {}
mRNA1_hmm = hmm.HMM.read(arg.work + 'C.elegans.mRNA1.hmm')
mRNA1_hmm.convert2log()
for dna in dnas:
mRNA1 = decode.Viterbi(model=mRNA1_hmm, dna=dna)
path = mRNA1.generate_path()
features = path.features()
for f in features: