logging.basicConfig(level=logging.INFO)
max_mosaics = 15
max_order = 5
me = evaluate_mosaics(max_mosaics=max_mosaics, max_order=max_order)
for i in range(max_order):
e=me[i*max_mosaics:(i+1)*max_mosaics]
plot([x[1] for x in e], [x[2] for x in e])
xlabel('# mosaics')
ylabel('LL')
title('Evaluation of mosaic models of various Markov orders')
savefig('mosaic-evaluation.png', format='PNG')
raise
# load our sequences
sequences = convert_fasta_sequences(fasta_file_for_fragment('T00671'))
# build our model
model_by_states = create_mosaic_model(
num_mosaics=1,
p_transition=0.,
alphabet_size=4,
order=2,
dirichlet_prior_strength=10.
)
model = hmm.as_model(model_by_states)
print model.B
# convert our sequences to the correct order
sequences_order_n = [model.converter.to_order_n(s) for s in sequences]
logging.basicConfig(level=logging.INFO)
max_mosaics = 15
max_order = 5
me = evaluate_mosaics(max_mosaics=max_mosaics, max_order=max_order)
for i in range(max_order):
e = me[i * max_mosaics:(i + 1) * max_mosaics]
plot([x[1] for x in e], [x[2] for x in e])
xlabel('# mosaics')
ylabel('LL')
title('Evaluation of mosaic models of various Markov orders')
savefig('mosaic-evaluation.png', format='PNG')
raise
# load our sequences
sequences = convert_fasta_sequences(fasta_file_for_fragment('T00671'))
# build our model
model_by_states = create_mosaic_model(num_mosaics=1,
p_transition=0.,
alphabet_size=4,
order=2,
dirichlet_prior_strength=10.)
model = hmm.as_model(model_by_states)
print model.B
# convert our sequences to the correct order
sequences_order_n = [model.converter.to_order_n(s) for s in sequences]
#from IPython.Debugger import Pdb; Pdb().set_trace()
def callback(LL):
# Copyright John Reid 2008
#
"""
Code to generate negative test sequences for those fragments in the test harness.
"""
from gapped_pssms.data import fasta_file_for_fragment, test_set_fragments
import sys
def sequences_from_fasta(fasta):
"""Yields sequences from fasta file."""
import corebio.seq_io.fasta_io
from itertools import imap
return imap(
lambda s: s.strip("nN"), imap(str, corebio.seq_io.fasta_io.iterseq(open(fasta, "r"), corebio.seq.dna_alphabet))
)
for fragment in test_set_fragments:
seqs = list(sequences_from_fasta(fasta_file_for_fragment(fragment)))
seq_length = max(len(s) for s in seqs)
num_seqs = len(seqs)
sys.argv = (
"generate_negative_test_sequences.py -m ..\..\Python\%s-bg-model.pickle -n %d -l %d -o negative-%s.fa"
% (fragment, num_seqs, seq_length, fragment)
).split()
execfile("generate_negative_test_sequences.py")
#
option_parser = OptionParser()
add_algorithm_options(option_parser)
logging.info('Command line: %s', ' '.join(sys.argv))
options, args = option_parser.parse_args()
log_filename = os.path.join(options.output_dir, '%s.log' % options.tag)
logging.getLogger('').addHandler(logging.FileHandler(log_filename))
logging.info('Writing log to %s', log_filename)
for option in option_parser.option_list:
if option.dest:
logging.info('%s: %s (%s)', option.dest,
str(getattr(options, option.dest)), option.help)
#inputs = [('K10-g0.50-N200-L200-seed4-1', fasta_file_for_synthetic_data('K10-g0.50-N200-L200-seed4-1'))]
inputs = [(fragment, fasta_file_for_fragment(fragment))
for fragment in test_set_fragments]
# for each input sequence
for seq_tag, fasta_file in inputs:
# add a file handler to log for this test set
file_handler = logging.FileHandler(
os.path.join(options.output_dir, '%s.log' % seq_tag))
logging.getLogger('').addHandler(file_handler)
try:
sequences = convert_fasta_sequences(fasta_file)
#sequences = [s[:200] for s in sequences[:10]]
# set up the options for this test set
options.tag = seq_tag
options.bg_model_filename = "%s-bg-model.pickle" % seq_tag
# Parse the options
#
option_parser = OptionParser()
add_algorithm_options(option_parser)
logging.info('Command line: %s', ' '.join(sys.argv))
options, args = option_parser.parse_args()
log_filename = os.path.join(options.output_dir, '%s.log' % options.tag)
logging.getLogger('').addHandler(logging.FileHandler(log_filename))
logging.info('Writing log to %s', log_filename)
for option in option_parser.option_list:
if option.dest:
logging.info('%s: %s (%s)', option.dest, str(getattr(options, option.dest)), option.help)
#inputs = [('K10-g0.50-N200-L200-seed4-1', fasta_file_for_synthetic_data('K10-g0.50-N200-L200-seed4-1'))]
inputs = [(fragment, fasta_file_for_fragment(fragment)) for fragment in test_set_fragments]
# for each input sequence
for seq_tag, fasta_file in inputs:
# add a file handler to log for this test set
file_handler = logging.FileHandler(os.path.join(options.output_dir, '%s.log' % seq_tag))
logging.getLogger('').addHandler(file_handler)
try:
sequences = convert_fasta_sequences(fasta_file)
#sequences = [s[:200] for s in sequences[:10]]
# set up the options for this test set
options.tag = seq_tag
options.bg_model_filename = "%s-bg-model.pickle" % seq_tag
# Run the algorithm
