donor0_data = classify(matrixDonor0, 2)
acceptor0_data = classify(matrixAcceptor0, 2)
no_coding_dist = calculator.intron_calculator('cuts_intron.txt').p
donor_states = sequence_state_factory(donor0_data, 'donor0')
acceptor_states = sequence_state_factory(acceptor0_data, 'acceptor0')
intron_spacer_states = spacer_states_maker(10, no_coding_dist, 'intron spacer')
utr_model = HiddenMarkovModel('utr_model')
# States
exon_state = State(DiscreteDistribution(calculator.utr_exon_5('mcutsa.txt').p),
name='utr exon')
intron_state = State(DiscreteDistribution(no_coding_dist), name='utr intron')
utr_model.add_model(promoter_model)
utr_model.add_state(exon_state)
utr_model.add_state(intron_state)
add_sequence(utr_model, donor_states)
add_sequence(utr_model, acceptor_states)
add_sequence(utr_model, intron_spacer_states)
utr_model.add_transition(utr_model.start, get_state(promoter_model, 'back'), 1)
utr_model.add_transition(get_state(promoter_model, 'inr7'), exon_state, 1)
utr_model.add_transition(get_state(promoter_model, 'no inr7'), exon_state, 1)
utr_model.add_transition(exon_state, exon_state, 0.7)
utr_model.add_transition(exon_state, donor_states[0], 0.2)
utr_model.add_transition(exon_state, utr_model.end, 0.1)
