def get_flying_sequence(flight):
state_alphabet = Alphabet()
state_alphabet.letters = list("fs")
emissions_alphabet = Alphabet()
emissions_alphabet.letters = list("FS")
emissions = []
for x in flight._flying_emissions():
if x == 1:
emissions.append("F")
else:
emissions.append("S")
emissions = Seq("".join(emissions), emissions_alphabet)
empty_states = Seq("", state_alphabet)
return TrainingSequence(emissions, empty_states)
def get_flying_sequence(flight):
state_alphabet = Alphabet()
state_alphabet.letters = list("fs")
emissions_alphabet = Alphabet()
emissions_alphabet.letters = list("FS")
emissions = []
for x in flight._flying_emissions():
if x == 1:
emissions.append("F")
else:
emissions.append("S")
emissions = Seq("".join(emissions), emissions_alphabet)
empty_states = Seq("", state_alphabet)
return TrainingSequence(emissions, empty_states)
def initial_markov_model_flying():
state_alphabet = Alphabet()
state_alphabet.letters = list("fs")
emissions_alphabet = Alphabet()
emissions_alphabet.letters = list("FS")
mmb = MarkovModelBuilder(state_alphabet, emissions_alphabet)
mmb.set_initial_probabilities({'f': 0.20, 's': 0.80})
mmb.allow_all_transitions()
mmb.set_transition_score('f', 'f', 0.99)
mmb.set_transition_score('f', 's', 0.01)
mmb.set_transition_score('s', 'f', 0.01)
mmb.set_transition_score('s', 's', 0.99)
mmb.set_emission_score('f', 'F', 0.90)
mmb.set_emission_score('f', 'S', 0.10)
mmb.set_emission_score('s', 'F', 0.10)
mmb.set_emission_score('s', 'S', 0.90)
mm = mmb.get_markov_model()
return mm
def initial_markov_model_flying():
state_alphabet = Alphabet()
state_alphabet.letters = list("fs")
emissions_alphabet = Alphabet()
emissions_alphabet.letters = list("FS")
mmb = MarkovModelBuilder(state_alphabet, emissions_alphabet)
mmb.set_initial_probabilities({'f': 0.20, 's': 0.80})
mmb.allow_all_transitions()
mmb.set_transition_score('f', 'f', 0.99)
mmb.set_transition_score('f', 's', 0.01)
mmb.set_transition_score('s', 'f', 0.01)
mmb.set_transition_score('s', 's', 0.99)
mmb.set_emission_score('f', 'F', 0.90)
mmb.set_emission_score('f', 'S', 0.10)
mmb.set_emission_score('s', 'F', 0.10)
mmb.set_emission_score('s', 'S', 0.90)
mm = mmb.get_markov_model()
return mm
#INPUT PHASE
lines = []
with open('input.txt','r') as handle:
for l in handle.readlines():
lines.append(l)
iterations = int(lines[0])
x = lines[2].strip()
alphabet = lines[4].split()
emission_alphabet = Alphabet()
emission_alphabet.size = 1
emission_alphabet.letters = alphabet
states = lines[6].split()
states_alphabet = Alphabet()
states_alphabet.size = 1
states_alphabet.letters = states
transition_probs = np.zeros((len(states),len(states)))
for i in range(len(states)):
prob = lines[9+i].split()
for j in range(len(states)):
put = prob[1+j]
transition_probs[i][j] = float(put)
emission_probs = np.zeros((len(states), len(alphabet)))
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("fasta")
parser.add_argument("slop", type=int)
parser.add_argument("flagged")
parser.add_argument("--primer3", action="store_true", help="output in primer3-compatible input format")
parser.add_argument("--min_primer_size", type=int, default=18)
parser.add_argument("--opt_primer_size", type=int, default=20)
parser.add_argument("--max_primer_size", type=int, default=27)
parser.add_argument("--max_primers_to_return", type=int, default=5)
args = parser.parse_args()
slop = args.slop
alphabet = Alphabet()
alphabet.letters = ["A", "T", "C", "G", "[", "]"]
with open(args.flagged, "w") as fh:
for seq_record in SeqIO.parse(args.fasta, "fasta"):
if args.primer3:
sequence = str(seq_record.seq)
fh.write("SEQUENCE_ID=%s\n" % seq_record.id)
fh.write("SEQUENCE_TEMPLATE=%s\n" % sequence)
fh.write("SEQUENCE_TARGET=%i,%i\n" % (slop, len(sequence) - 2 * slop))
fh.write("PRIMER_OPT_SIZE=%i\n" % args.opt_primer_size)
fh.write("PRIMER_MIN_SIZE=%i\n" % args.min_primer_size)
fh.write("PRIMER_MAX_SIZE=%i\n" % args.max_primer_size)
fh.write("PRIMER_NUM_RETURN=%i\n" % args.max_primers_to_return)
fh.write("=\n")
else:
sequence = Seq(
